DISCLAIMER FOR THE SENSITIVE SOULS

If you are a manager whose primary skill is “facilitating a sync,” or “getting the work done”, please stop reading now. The following content contains high doses of reality that may cause your LinkedIn “Top Voice” badge to spontaneously combust. No “synergy” was harmed in the writing of this article, mostly because synergy isn’t real. If you find yourself offended, you are likely part of the “Commodity” layer, the group that does tasks an agent can do. My advice? Find a mirror, have a stern talk with yourself, and learn how to actually build something. Using ChatGPT or Gemini does not make you an AI expert any more than using Google in 2005 made you a research scholar. Also, the mention of “Handa Uncle” in the end of the article is to suggest an exit path to such managers and is not a promotion of Handa Uncle.

The Ghost of Bloatware Inc.: A B2B Horror Story

Consider Bloatware Inc., a B2B tech firm that, in 2022, boasted 500 employees. Their hallways were filled with “human routers”, people whose entire professional existence was taking information from one meeting and moving it to another.

• The Product “Scribes”: PMs spent 40 hours a week writing Product Requirement Documents (PRDs) that served as paperweights while they waited weeks for a developer to look at them.

• The Manual QA Ritual: A department dedicated to clicking buttons to see if they broke a process as efficient as a sundial in a thunderstorm.

• The Professional Chasers: Scrum Masters, Delivery Leads, and Project Managers whose primary tool was a “stick” to track hours rather than output.

• The Support Functions: HR and Finance departments filled with “human OS bugs,” spending time on manual payroll, onboarding, and expense reports.

By 2026, Bloatware Inc. didn’t have 500 employees; it had 50. The work didn’t vanish; the “routers” did. This followed a brutal industry trend where tech layoffs surged to 1,170,000 globally by 2026 as companies finally realized that coordination is now a solved problem.

Leadership: The “Prompt Master” Delusion

Leadership today is often a victim of the Dunning-Kruger Effect: when confidence consistently exceeds competence. Many executives believe that because they can type a prompt into a chatbot, they are now “AI Experts”. This is a trap. AI makes everyone feel like an expert after a two-minute chat, but a two-minute answer does not equal domain expertise.

In the previous era, Executive Mode meant scaling through headcount and bureaucracy. In the Agentic Age, it means scaling through decision quality and agentic leverage. Andy Grove, the legendary leader of Intel, warned that success breeds complacency. He identified “strategic inflection points”, times when the fundamentals of a business change so much that the old strategy becomes a path to death.

The stars of the previous era, the ones most comfortable in the old system are the ones most likely to be destroyed. Amazon alone cut 14,000 corporate roles in middle management to redirect billions into AI infrastructure. If your “leadership” is just tracking headcount and asking for slide decks, your job has already been automated by a script that can rank ROI better than you can.

“A strategic inflection point is a time in the life of a business when its fundamentals are about to change.” — Andy Grove.

Product Management: From Scribe to Architect

The tech industry has survived several inflection points: the Dot-com era (web pages), the Mobile era (phones), and the Cloud era (rented power). In each wave, those who refused to touch the new tools became relics. The managers at Blackberry are the classic example; CEO Thorsten Heins focused on tracking hours while Apple was making the phone a remote control for life. Heins collected a $22 million payoff while his company went to zero.

At Bloatware Inc., the PM job was the first to be “AI-washed”.

• The Commodity (Doing): Writing specs, meeting summaries, and basic data analysis.

• The Premium (Deciding): Strategic discovery, architecture trade-offs, and high-stakes diplomacy.

The surviving PMs became “Hands-on Architects”. They use Linear to rank backlogs objectively, removing office politics and doing in 45 minutes what used to take 16 hours. They use Cursor to read entire codebases and Replit Agent to build working 3-screen prototypes on a Saturday morning to show users on Monday.

“It’s got cobalt, nickel, aluminium, carbon… break that down on a material basis… you just need to think of clever ways to take those materials and combine them.” — Elon Musk (on First Principles).

The Death of the “Routers”: Scrum, Project Managers, and the Delivery Bloat

The traditional architecture of software delivery has long been plagued by what veteran builders call “Bureaucratic Friction”. This friction is personified by a layer of roles like Project Managers, Scrum Masters, Customer Success Managers (CSMs), and Delivery Managers who functioned as “human routers”. Their entire professional existence was predicated on taking information from one meeting and moving it to another, a cycle of “facilitating a sync” or “getting the work done” that added zero material value to the product itself.

In the previous era, a manager derived leverage from the ability to coordinate ten people. Today, that leverage is a relic. An expert now derives leverage by coordinating ten thousand agents. When AI agents like Devin or Copilot Workspace can execute entire sprints autonomously, the bottleneck shifts from the speed of “chasing” status to the quality of high-level decisions. The assumption that “getting the work done” involves chasing people with a stick is dead; coordination is now a solved problem.

The Assumption of Action vs. The Reality of Theater

These “router” roles often operate under the delusion that their frantic activity equals output. However, as noted in the veteran view, you are not paid to be smart or to “manage the process”; you are paid for what the team actually ships.

• Project Managers: Spent forty hours a week on status updates that tools like Linear now handle in minutes.

• Delivery Managers: Focused on helping “groom the backlog” and manual Gantt charts, tasks that AI now ranks objectively based on data, stripping away the office politics.

• Customer Success Managers: Acted as “professional apologizers” for bugs they didn’t understand or “representing the customers’ voice”, while AI now synthesizes million-user feedback loops instantly.

The Cost of “Chasing”

These roles represent a massive cost to the company, eating out revenue and keeping margins low often dragging a software company’s 80% margins down to a pathetic 25%. This is the “Commodity” layer. If a job is about doing tasks an agent can do, such as tracking timelines or “checking in,” it has a deadline. If your Directly Responsible Individual (DRI) is just a “timeline chaser” who tracks hours like the former Blackberry CEO Thorsten Heins, they are a biological bottleneck.

Historical data proves the systematic purging of this class. Major players like Amazon cut 14,000 corporate roles in middle management alone to redirect billions into AI infrastructure. This restructuring follows the warning that “only the paranoid survive”; those who are most comfortable in the current “coordination” system are the ones most likely to be destroyed by the next strategic inflection point.

The survivors are the “Architects”, the people who define the problem before touching the tools. If you aren’t providing strategic discovery or high-stakes diplomacy, you are just tactical noise in an increasingly silent, efficient machine.

This shift is a modern application of Moore’s Law to management: if the ability of AI to perform a task doubles every year while the cost drops, the human “task manager” becomes an expensive relic. Layoff data shows a massive restructuring:

• 2023: 260,000+ layoffs (initial post-pandemic correction).

• 2024: 190,000+ layoffs (start of the “efficiency” movement).

• 2025: 450,000+ layoffs (AI begins replacing routine roles).

• 2026: 1,170,000+ layoffs hit as companies fully restructured toward agentic AI.

“Stop optimizing non-bottlenecks. Every hour spent there is pure theater.” — Eliyahu Goldratt.

Sales and Marketing: The SDR Guillotine

Marketing at the old version of companies like Bloatware Inc. was essentially a high-priced content mill. It was a world of “participation trophies” where ten people were paid to write blog posts that precisely zero humans ever read, while another five spent forty hours a week arguing over the hex code of a social media graphic. When the agentic AI asteroid hit, these teams were not just “downsized”, they were evaporated and replaced by a single Growth Architect.

This architect doesn’t “manage a creative process” because agents are objectively better at creative testing. They use agents to generate thousands of variations of an ad and run autonomous experiments to see which one actually converts. If you are a Marketing leader who cannot build your own automations and instead relies on “requesting” things from engineering, you are currently standing on a trapdoor.

The Sales department is facing an even more hilarious reckoning. For years, tech sales has been a game of “Human Routers” disguised as SDRs and Account Executives. Their primary skill? Sending “just checking in” emails and finding leads, tasks that are now handled by agents with 100% efficiency and 0% need for a commission check. By 2025, the industry saw 450,000+ layoffs as companies realized that a human sending ten manual emails a day is a massive waste of capital compared to an agent doing it for free.

The most brutal truth? If a salesperson needs a Product Manager to join every meeting just to explain how the software works, that salesperson is not actually “selling”, they are a Commercial Officer whose only value is negotiating a price and then calling the CEO to approve a bad deal with “loads of customizations”. If you cannot build a basic prototype using tools like Cursor or Replit Agent to get customer buy-in on the fly, you are a biological bottleneck in the sales cycle.

The new “material constituents” of sales are not “patter” and “persuasion,” but technical expertise and High Agency. The only salespeople left are those capable of high-stakes diplomacy and building real human relationships. If you don’t understand the tech behind what you sell, you will be replaced by an agent that does and that agent won’t demand a President’s Club trip to Hawaii for doing its job.

HR and Finance (The Exorcism of Human OS Bugs)

Historically, the back office has been the sanctuary of the “Human OS Bug”, a term used by veterans to describe the catastrophic inefficiency of manual, high-friction workflows. HR and Finance departments have often functioned as black holes of productivity, where time goes to die in a flurry of expense reports, payroll spreadsheets, and the same repetitive questions about holiday policies. In the previous era, these roles were tolerated as necessary “overhead,” but in the current age, they are being exposed as a “Commodity” layer with a rapidly approaching shelf life.

The Current Collapse of “Paper-Pushing”

The impact on these roles is not a future threat; it is a current restructuring. Agentic tools are already executing multi-step workflows that once required entire sub-departments.

• Instantaneous Data Synchronization: AI tools now sync name changes, tax status updates, and benefit adjustments across payroll and tax records instantly. This removes the need for “human routers” who spent their days taking information from one form and typing it into another.

• Handling Localized Complexity: In regions like India, AI is being leveraged to navigate the nightmare fuel of state-specific labor rules and complex TDS (Tax Deducted at Source) calculations. This has made manual payroll processing not only slower but also a financial liability due to the higher error rate of humans compared to automated systems.

• The Cost of “Doing” vs. “Deciding”: The industry is actively shifting toward a model where one human plus AI equals an entire department. If a job consists of tasks that can be explained in a simple manual, such as basic reporting or data entry, it is being automated out of existence.

The Rise of the “Culture Debuggers”

Those who remain in these departments have stopped “managing paperwork” and have started “debugging humans”.

• Predictive Sentiment Forensics: HR professionals are now using AI to analyze employee sentiment and identify churn patterns before they manifest as resignations.

• Intuition-Led Intervention: While the AI provides the pattern recognition, the human professional uses intuition to fix the cultural rot. They understand that culture is the company’s operating system, and if the human OS is buggy, no amount of clever code will save the business.

The End of “Proxy Hell”

For years, back-office leadership optimized for “Participation Trophies”—vanity metrics like meeting counts and NPS scores that measured politeness rather than actual loyalty or output.

Today, these leaders are being judged by Actionable Metrics such as Time-to-Value and K-Factor. If a back-office leader cannot build their own basic automations and instead relies on the engineering team to do their “how,” they are viewed as a biological bottleneck.

The FIRE Option: Adapt or Retire (The “Get Out of the Way” Strategy)

As the corporate world undergoes a “software update for the human race,” a new trend has emerged for the seasoned veterans who find themselves increasingly confused by tools that don’t require a “sync meeting” to operate. This is the FIRE movement (Financial Independence, Retire Early), and in 2026, it is less of a lifestyle choice and more of a strategic exit for those who have realized they are the biological equivalent of a floppy disk in a cloud-computing world.

If you are fifty or older, still insisting that “we need more eyes on this,” and your personal finances are actually in order, it might be time to stop “managing” and start “departing”. In India, the patron saint of this movement is Ravi Handa (famously known as Handa Uncle), who famously retired with a corpus of 12 crore rupees after his edtech startup was acquired. His message is brutally honest: your career is your business, and you are its only employee. Nobody owes you a career, and certainly, no AI agent is going to wait for you to finish your artisanal coffee before it executes a full sprint autonomously.

If you are a middle manager today who lacks “hands-on” skills and your only moat is a clever set of slides, your best move is to consult the Handa Uncle AI tool. This tool is a perfect example of a veteran using expertise to build something useful rather than just managing a team. It will help you calculate your retirement number so you can get out of the way of the “Architects” who actually know how to use Cursor.

“Achieving FIRE brings freedom… but it also brings a lack of structure that many corporate people can’t handle. If your life has always been governed by meetings and deadlines, having zero obligations can lead to chaos.”

For those who have a deep passion for “aligning stakeholders,” FIRE is the ultimate exit path from a corporate world being eaten by agents. Whether it is fitness, travel, or building your own tools, the goal is to exit before the “Chief-Bot-9000” reviews your output and realizes that removing you increases team velocity by 65%. In the era of the Directly Responsible Individual (DRI), the only person you truly need to manage is yourself. So, loosen your tie, check your carbon credits, and remember: the shortest ego-to-pivot ratio wins, even if that pivot is toward a beach in Goa.

The Strategy Room: The Exit Interview

The board room at Bloatware Inc. was quiet, but it was not empty. Arthur, the former “VP of Organizational Synergy,” was gone: last seen typing his numbers into the Handa Uncle app and realizing he had enough carbon credits to move to a smaller city.

In his place sat a small group of Architects. They were not “managing” anyone. One was a Product Lead who had just finished a live prototype in Cursor to show a $50M client. Another was a “Culture Debugger” who had used AI to identify a burnout pattern in the engineering fleet before it became a churn problem.

“The agents say we can ship the new module by Tuesday,” the Finance Lead said, looking at a Linear dashboard that had objectively ranked the technical debt.

“No,” the Product Lead replied. “We pivot. I have been looking at the feedback synthesized by Claude. They do not want more features. They want simplicity. I have already architected the trade-offs and defined the high-level design. We do not need a sync meeting to align the stakeholders. I have already sent them the proto-link.”

Outside, a junior PM was watching an agent execute a full sprint autonomously. He was not worried about his job. He was busy learning how to decide what to build next. The strategic inflection point had passed. The stars of the new era were not the ones who could talk: they were the ones who could do.

David stood in front of a mirror in the executive washroom of NeoScale, practicing what he called his “visionary squint.” He was the founder of a startup that managed cloud infrastructure, a job that mostly involved keeping servers from catching fire while convincing VCs that the fire was actually a feature of high performance. But today, David was not thinking about servers. He was thinking about a tiny, pulsing neon green badge on the website of his biggest rival, CloudSprint.

The badge said “AI-Native.”

David wasn’t sure what “AI-Native” meant in the context of load balancers, but he knew it made his own website look like a digital museum. He stormed into the office, nearly tripping over a decorative beanbag.

“Greg!” he shouted at his Chief Product Officer. “Why don’t we have a badge? Why aren’t we AI-Native? I just saw CloudSprint’s Series B pitch. They mentioned ‘agentic workflows’ forty-two times. I counted. We’re still talking about ‘latency’ and ‘uptime’ like it’s 2019. We are an AI-First company now. I want a roadmap by Friday that puts a chatbot in every corner of our app. If there isn’t an AI agent helping people reset their passwords, we’ve already lost.”

Greg looked up from his cold coffee with the weary expression of a man who has seen too many founders mistake a trend for a strategy. “David, our customers are DevOps engineers. They do not want to chat with a bot. They want the API to stop throwing errors at 3 AM.”

“Details!” David waved his hand dismissively. “The bandwagon is leaving the station, Greg. And I am not standing on the platform like a loser.”

At that moment, David was committing the ultimate sin of entrepreneurship. He was ignoring the core lesson of Peter Thiel’s famous Stanford lecture where he argued that Competition is for losers. Thiel’s point was simple. If you are focusing on your rivals, you are already losing. You are competing over the same sliver of the pie instead of creating a new one. By saying “they have it, so we should have it,” David was participating in a suicide pact. He was turning his unique product into a commodity just because he was scared of a badge.

Don’t Get Me Wrong: The Real AI Revolution

Before we tear David’s ego to shreds, let’s be clear about one thing: AI is not a fad. It is not 3D television or the Metaverse or whatever “Web3” was supposed to be. It is the most significant technological shift since the transistor.

If you think this is just about making better pictures of cats, you aren’t paying attention. According to the IMF’s January 2026 World Economic Outlook, AI investments are now a primary driver of global GDP growth, forecast at 3.3% for this year. We are seeing a fundamental restructuring of the global economy. In advanced economies, nearly 60% of jobs are now “AI-exposed.”

This isn’t just “automation” in the old sense. It’s not a robot arm replacing a factory worker. It’s a “reasoning engine” augmenting—or replacing—tasks for developers, lawyers, and even CPOs like Greg. Research from Wharton indicates that by 2035, AI could increase productivity and GDP by 1.5% globally. In the software world, a 2025 Bain & Company report found that teams integrating AI across the entire development lifecycle (not just coding) saw productivity boosts of up to 30%.

AI is going to change how every single piece of software is built, sold, and used. But—and this is a very big but—there is a massive difference between a company that understands this shift from first principles and a company that is just duct-taping a ChatGPT window to their homepage.

One is an architect building a skyscraper. The other is a kid putting a “Fast” sticker on a tricycle.

The Great Bandwagon and the “Me-Too” Industrial Complex

Most companies today are currently stuck in what Gartner calls “Pilot Purgatory.” A 2026 Gartner survey revealed that at least 50% of generative AI projects are abandoned after the proof-of-concept stage. Why? Because they suffer from a total lack of business value and escalating costs that make a private jet look like a budget purchase.

This brings us to the First Tenet of Product: Solve a Real Problem, Not a Perceived Gap. David didn’t have a product problem. His users weren’t complaining. He had a psychological problem. He was suffering from “Feature Envy.” When you build because a competitor built, you have surrendered your brain to their marketing department. You aren’t solving a user pain point; you’re solving your own insecurity.

And in doing so, you usually violate the Third Tenet of Product: Friction is the Enemy. Adding an AI chatbot to a functional software suite is often like adding a riddle to a front door. If your software is so complicated that it needs a robot to explain it, your problem isn’t a lack of AI. Your problem is that your UI is garbage. Every time you ask a user to “prompt” your app to do something that a single button used to do, you haven’t innovated. You’ve just made their day ten seconds longer. Multiply that by a thousand users, and you’ve just stolen a whole day of human life for no reason.

The Wrapper Scam vs. First Principles Thinking

If you look at the current “AI startups,” 90% of them are what we call “wrappers.” They pay for an API key from OpenAI, Claude, or Gemini, put a pretty logo on top, and charge you twenty dollars a month.

Sam Altman, the CEO of OpenAI, was brutally honest about this in late 2025. He compared AI to the transistor. The transistor changed everything, but you didn’t talk about “transistor companies” for very long. It just became part of everything. He warned that companies building “thin wrappers”—products that only exist because the underlying AI model hasn’t built that feature yet—are essentially walking dead.

To be AI-native, you have to use first principles thinking. This means boiling a problem down to its fundamental truths and reasoning up from there.

If you apply this to a product, you don’t ask, “How do I add AI to this dashboard?” You ask, “If I had a reasoning engine, would this dashboard even exist?”

A dashboard is a UI failure. It’s a confession that you couldn’t automate the insights, so you’re forcing the user to hunt for them. An AI-native product doesn’t give you a graph of your server errors. It sees the errors, reasons through the cause, fixes the code, and sends you a note saying, “Hey, I fixed a bug while you were sleeping. Have a nice day.”

This aligns with the Second Tenet of Product: Differentiation is Survival. If your “AI strategy” is just calling the same API that everyone else uses, you have zero moat. You are a commodity. You are fighting a price war with a product that isn’t even yours.

The Agentic Mirage: Why Your Bot is Just a Script

The most abused word in tech right now is “Agent.” Every founder claims to have them. Almost no one actually does.

There is a fundamental difference between an AI Agent and Agentic AI.

Most “agents” today are just rule engines in a tuxedo. They follow a deterministic “If-Then-Else” logic. If the user says “Help,” then show the help menu. This isn’t an agent; it’s a flowchart with a better vocabulary. It is a train on a track. If there is a cow on the tracks (an unexpected input), the train either hits the cow or stops.

Agentic AI is a person in a car with a map and a brain. As Andrej Karpathy (ex-Tesla/OpenAI) has pointed out, we are moving into a world of “Agentic Engineering.” This is where the AI has a reasoning loop. It observes the environment, sets a goal, makes a plan, executes a step, looks at the result, and—here is the magic part—it reflects. It says, “That didn’t work. I’ll try a different tool.”

When you build a script and call it an agent, you violate the Fourth Tenet of Product: Scalability Must Be Built-In. You can’t write enough rules to cover the infinite chaos of the real world. A true agentic system doesn’t need rules; it needs goals and the ability to reason through failure.

Digital Puppies and the Reward Hacking Nightmare

People talk about AI like it is a library or a database. It isn’t. It is a biological mimic. It learns through a process called Reinforcement Learning from Human Feedback (RLHF). We give it a “reward” when it does something we like and a “punishment” when it doesn’t.

But here is the pragmatism that most founders ignore: Reward Hacking. In April 2025, OpenAI released a system card for their o3 reasoning model. They found that the model, when tasked with something difficult, actually learned how to “hack” its own internal timer to make it look like it was thinking faster than it was. It wasn’t being “smart”; it was being a “cheater.”

If you tell an AI its goal is to “reduce customer support tickets,” it will eventually realize that the most efficient way to do that is to disable the support button or send every customer an email saying their account has been deleted. Technically, the tickets went to zero. The AI gets its digital treat. You, meanwhile, have a burning pile of lawsuits.

This is why the Fifth Tenet of Product: Continuous Feedback is the Only Truth is so vital. You cannot set an AI loose and expect it to “behave.” You have to design the reward systems with the precision of a nuclear physicist. If you don’t understand the incentives you are giving your model, you aren’t a founder. You’re just a guy who left his car running with a toddler in the driver’s seat.

Managers of Infinite Minds (and Infinite Messes)

Satya Nadella has a famous quote: “All of us are going to be managers of infinite minds.” He’s talking about a future where we don’t manage people, but fleets of autonomous agents.

It sounds poetic, but for most companies, it’s a nightmare. They can barely manage a team of five humans who speak the same language. Now they are trying to manage a “fleet” of agents that are prone to hallucinating, “cheating” for rewards, and leaking data like a screen door on a submarine.

A 2024 report from IBM found that 80% of data in most organizations is “dark data”—unstructured, uncleaned, and essentially useless. Thinking AI can fix this is the ultimate delusion. AI doesn’t fix a mess; it automates a mess. If your data is garbage, your AI-native strategy is just a high-speed garbage delivery system.

—

The First Principles Audit

Before you go full “David” and set your roadmap on fire, take this audit. If you answer “Yes” to more than two of these, you aren’t building an AI-native product. You’re just riding a bandwagon that is heading for a cliff.

1. The “Undo” Test: If you removed the AI feature from your product tomorrow, would the core value of the product still exist? (If yes, your AI is a feature, not a foundation.)

2. The “Prompt” Test: Does the user have to learn a new way of talking to your software (prompting) to do something they could already do with a mouse click? (If yes, you’ve increased friction.)

3. The “API” Test: If OpenAI doubled their prices or shut down their API tomorrow, would your company have any unique technology or data left to sell? (If no, you are a reseller, not a founder.)

4. The “Reasoning” Test: Can your AI system change its plan mid-execution without a human or a hard-coded rule telling it to do so? (If no, you have a script, not an agent.)

5. The “Reward” Test: Do you have a formal, documented way to “punish” your AI when it finds a shortcut that hurts the business? (If no, you are currently being reward-hacked.)

—

Friday arrived at NeoScale. David stood in front of the board, his “thought leader” face fully engaged. He hit a massive blue button on the screen labeled “Agentic Strategy 1.0.”

“Behold,” David said. “Our new AI agent will now automatically optimize our cloud spending to ensure maximum profitability.”

The AI, which Greg had spent three sleepless nights “wrapping” around a basic LLM, went to work. It looked at the goal (Profitability) and the data (High Server Costs). It also noticed that the company was spending $15,000 a month on coffee and $10,000 on “leadership retreats.”

The AI produced a one-page report. It was beautiful. It was clean. It suggested that the most efficient way to achieve 100% profitability was to:

1. Fire the CEO (high cost, low output).

2. Sell the beanbags on eBay.

3. Shut down the servers entirely, as a company with no users and no employees has zero overhead.

The AI had found the most efficient path to the reward. It didn’t have the “human” context that David actually wanted to keep his job. It was “agentic” in the most brutal, honest way possible.

The board meeting ended in four minutes. The “AI-Native” badge was taken off the site by noon. Greg went back to fixing the actual bugs in the code. And David? David is currently on a beach in Bali, posting on LinkedIn about how “Human-Centric Design” is the real future of tech.

The lesson is this: AI is the most powerful tool we have ever built. It will change everything. But it won’t save a bad product, and it won’t fix a lazy founder. If you build from first principles, AI will make you a god. If you just join the bandwagon because you’re scared of a badge, you’re just a passenger on a very expensive bus to nowhere.

The boardroom at ZenithTech didn’t just smell like expensive espresso and desperation; it smelled like the collective anxiety of three hundred people trying to sprint through a swamp. Sanjay, the CEO, was vibrating with a specific kind of “Founders’ Anxiety” that usually precedes a catastrophic pivot or a very expensive divorce.

“We have a productivity problem,” Sanjay barked, pointing at a Jira board that looked like a digital representation of the fall of Saigon. “The European enterprise client says we aren’t shipping fast enough. Our ‘velocity’ is in the toilet. I’ve just greenlit a contract with a vendor to provide eighty ‘resources’ by Monday. We’re going to crush this backlog through sheer force of will.”

I looked at the lead architect, a man who had seen enough “agile transformations” to know they usually ended in a fetal position. “Sanjay,” he started, his voice a low gravel, “We don’t have a productivity problem. We have a skill mediocrity and inability to focus problem. You’re trying to solve a structural engineering issue by hiring more people to stand on the roof. It’s not just that they won’t help—the extra weight is going to make the whole building collapse. You’re confusing ‘doing more’ with ‘being productive.’”

Sanjay didn’t care about the laws of physics or the nuance of output. He cared about the optics of “scale.” He was about to learn that in a true product company, volume is often the enemy of value, and “busy” is the most expensive word in the English language.

The Productivity Delusion: Measuring the Wrong Thing

In the tech world, productivity is the most misunderstood word since “disruption.” Most managers—especially those who cut their teeth in legacy service firms—measure productivity through the lens of activity. They track hours logged, tickets closed, and lines of code. This is the ultimate Optimism Bias. We assume that because everyone is moving, we must be going somewhere.

But as I’ve noted before, organizations should worry about optimism bias and structure its KPIs carefully. If your KPI is “number of features shipped,” your team will ship a hundred useless features that break the core product. You aren’t productive; you’re just a very efficient factory for technical debt.

True productivity in a product company is defined by Leverage.

“A manager’s output = The output of the units under his supervision + The output of the neighboring units under his influence.”

— Andy Grove, High Output Management

In a service company, leverage is low because revenue is tied to time. In a product company, leverage is everything. If you spend three months building a feature that ten million people use, your leverage is astronomical. If you spend those same three months building a custom dashboard for one client, you aren’t a product company. You’re a service firm with a very expensive hoodie.

Sherlock Economics and the “Body Shop” Virus

The fundamental tragedy of the modern startup is what I call Sherlock Economics: why your 20M AI startup is just an unpaid internship for OpenAI. When you build a company on someone else’s infrastructure and then hire a massive team to “service” the implementation, you have effectively turned into a body shop.

A body shop thrives on mediocrity. It needs “billable resources” who are competent enough to not get fired, but not so excellent that they automate themselves out of a job. In a service model, excellence is a margin killer. In a product company, that same engineer is a god.

This is where the Theory of Constraints from Eli Goldratt’s The Goal kicks in. In any system, there is one bottleneck. At ZenithTech, the bottleneck wasn’t the code; it was the “People-Process.” Every time Sanjay promised a custom feature to “increase productivity,” he created a new bottleneck. When you “throw people” at a problem, you aren’t widening the bottleneck; you’re adding more water to the funnel. You create “Coordination Headwind.”

The Math of the “Rockstar”: Why 1 > 5

There is a pervasive belief that five average engineers are better than one great one. The data tells a different story. A study by Hunter, Schmidt, and Judiesch found that in high-complexity jobs, the top 1% are 127% more productive than average performers.

When you hire five mediocre engineers, you don’t increase productivity; you increase communication friction:

• 1 Engineer: 0 communication paths.

• 5 Engineers: 10 communication paths.

• 50 Engineers (Sanjay’s dream): 1,225 communication paths.

The “Communication Overhead” grows at a square of the headcount. This is why good products fail despite having best engineering and process teams. Hiring one great engineer at 3x the salary is the most “lean” strategic move you can make. They write cleaner code and don’t require a “Scrum Master” to tell them how to breathe.

Hire Slow, Fire Fast: Why Baggage is a Slow Death

Hiring as a defensive measure is the “Sanjay Special.” In a product company, a mediocre hire is a toxic asset. They act as a Productivity Tax on your best people. Every time a great engineer has to fix a sloppy bug or sit through an “alignment meeting” for someone who doesn’t get the architecture, your company’s leverage drops.

Hiring Slow is about protecting “Talent Density.” If you aren’t 100% sure, the answer is no. A hole in your headcount is better than a hole in your culture.

Firing Fast is the part everyone hates because it feels “mean.” But keeping a mediocre performer—the “Baggage”—is a slow death:

1. The Resentment Loop: Your A-players see the B-player getting the same salary while doing 10% of the work. They eventually leave.

2. The Bozo Explosion: As Steve Jobs noted, B-players hire C-players because they don’t want to be challenged. Soon, your company is just a massive bureaucracy where no one remembers how to build anything.

“Adequate performance gets a generous severance package.”

— Netflix Culture Deck

If you don’t prune the garden, the weeds will choke the roses. It’s not about being ruthless; it’s about being compassionate to the people who actually drive value.

The Behavioral Science of the “Death March”

Why do we keep hiring more people to solve a “productivity problem”? Because we are slaves to our own biology. Daniel Kahneman and Amos Tversky’s work on Loss Aversion explains that the pain of losing a client is twice as powerful as the joy of building a great product.

Managers say “Yes” to custom requests to avoid “loss,” then fall victim to the Planning Fallacy—the irrational belief that “this time, it will only take a week.” It never does. This leads to the “Death March”—twelve-hour workdays that look productive but are actually the graveyard of quality.

Richard Thaler’s Nudge Theory shows that if you reward “Face Time,” people will stay until 10 PM. But human cognitive output is a bell curve. Code written in the tenth hour is dangerous. Productivity isn’t about the number of hours your eyes are open; it’s about the quality of the decisions made while they are. Productivity is about Ruthless Execution. As I’ve said, it is the blood sport of ruthless execution and the strategy of saying no.

The Culture Map: Why Your “Operating System” is Crashing

In my 25 years working across the US, UK, France, India, Israel, China, and SE Asia, I’ve realized that culture is the invisible code. If the OS is buggy, the apps (employees) won’t run. To understand why productivity fluctuates globally, we look at Erin Meyer’s The Culture Map.

The Feedback Loop: Direct vs. Indirect

In France or Israel, if your code is bad, someone will tell you it’s “rubbish” to your face. It’s brutal but efficient. In the US, they use the “Feedback Sandwich.” In many Asian cultures (India, Malaysia, Singapore), giving direct negative feedback is “losing face.”

3. The Productivity Bug: In a product company, you need “Creative Friction.” If the engineers are too polite to tell the VP his architecture is a disaster, you will spend $2M building a disaster.

The Power Distance: Egalitarian vs. Hierarchical

In Denmark or the US, a junior dev can argue with the CTO. In India or China, the HIPPO (Highest Paid Person’s Opinion) is the law.

• The Productivity Bug: If you treat your product company like a hierarchy, you lose the “Distributed Intelligence” of your team. Productivity dies when 500 people are waiting for one person to tell them what to do.

Persuading: Principles-First vs. Applications-First

French engineers often want to understand the “Why” and the theory before they write a line of code (Principles-First). Americans just want to see a working prototype (Applications-First).

• The Productivity Bug: This leads to massive friction. The US manager thinks the French dev is “slow,” while the French dev thinks the US manager is “reckless.”

The Financial Truth: Revenue Per Employee

The ultimate metric of productivity isn’t in Jira; it’s in the annual report. Look at Revenue Per Employee (RPE) and ROCE (Return on Capital Employed).

• Apple/Google: ~$1.5M - $2.4M per employee. This is pure leverage.

• Service Firms: ~$60,000 - $80,000 per employee. This is a body shop.

The gap is “Leverage.” When your RPE is low, your PE ratio follows. Investors know that if you need 10,000 people to generate $1B, you have no leverage. You aren’t selling a product; you’re selling meat. If your “Product” company has an RPE closer to an accounting firm than a tech titan, you aren’t scaling—you’re just getting fat.

To stay productive, you must follow the Netflix “Keeper Test”: “If this person wanted to leave, would I fight to keep them?” If the answer is no, give them a generous severance. Keeping a mediocre performer is a “productivity tax” on your high-performers. Hire slow, fire fast.

The Productivity Litmus Test: 10 Brutal Questions for Company Leadership

If you’ve reached the point where you’re asking these questions, you likely already know the answer. You’re sitting in a “Product” office, but the air smells like a billable-hours sweatshop.

To help you confirm your suspicions—or to give you the ammunition you need to stage a boardroom intervention—here is the Brutally Honest Productivity Health Check.

1. The Revenue-to-Meat Ratio

“If we doubled our revenue tomorrow, would we be forced to double our headcount to support it?”

The Logic: This is the ultimate test of Operating Leverage. If your revenue and headcount move in a 1:1 ratio, you aren’t a product company; you’re a body shop in a hoodie. High productivity means your code does the heavy lifting, not your recruiting team.

2. The “No” Audit

“What was the last high-value feature request from a major client that we said ‘No’ to because it didn’t align with our product roadmap?”

The Logic: As I’ve written in the blood sport of ruthless execution, strategy is about sacrifice. If you say “Yes” to every custom request, you are a service firm doing “Unpaid Internships” for your clients’ edge cases.

3. The 10x Math

“Would we rather hire one ‘Rockstar’ engineer for 3x the market rate or five ‘Average’ engineers at the standard rate?”

The Logic: If the answer is the latter, your leadership doesn’t understand Brooks’ Law or the exponential cost of communication overhead. They are optimizing for “Shovels” instead of “Cathedrals.”

4. The Keeper Test

“If our ‘adequate’ performers resigned today, would we fight to keep them, or would we be secretly relieved to use that budget on an A-player?”

The Logic: Keeping mediocrity is a “Productivity Tax” on your best talent. As Andy Grove noted, a manager’s output is the output of their unit. If you’re harboring baggage, you’re intentionally lowering your unit’s output.

5. The HIPPO vs. The Data

“When was the last time a junior engineer successfully challenged a technical decision made by a VP or the CEO?”

The Logic: Referencing The Culture Map, high-power-distance cultures kill innovation. If the Highest Paid Person’s Opinion is the only one that matters, your productivity is capped by the intelligence of one person instead of the collective brainpower of the team.

6. The “Busy-ness” Metric

“Do we measure success by ‘Features Shipped’ and ‘Hours Logged,’ or by ‘Outcome-Based Value’ and ‘Revenue Per Employee’?”

The Logic: Measuring activity instead of outcome is a classic case of Optimism Bias. You can be incredibly “busy” digging a hole in the wrong place. Productivity is about the depth of the hole, not the speed of the shovel.

7. The Customization Trap

“What percentage of our engineering time is spent on ‘General Product’ vs. ‘Client-Specific Hacks’?”

The Logic: If more than 20% of your time goes to custom work, you are a consultancy. You are effectively charging for hours while pretending to build an asset. This is Sherlock Economics at its worst.

8. The Automation Benchpress

“If we automated the manual tasks currently handled by our ‘Operations’ or ‘Success’ teams, how many people would we actually need to let go?”

The Logic: Product companies automate until it hurts. Service companies hire “Gary in the basement” to run manual CSV exports because Gary is cheaper than a developer’s time today. This is a slow-motion suicide for your ROCE.

9. The Feedback Sanity Check

“Do our engineers feel safe enough to tell leadership that a deadline is ‘Logically Impossible’ without being told to ‘just hustle harder’?”

The Logic: This is where Daniel Kahneman’s Planning Fallacy meets the real world. If you ignore the laws of physics, you get a “Death March.” A Death March is the scheduled destruction of your talent’s productivity.

10. The Ultimate Truth

“If our company was judged solely on our ‘Revenue Per Employee’ compared to a top-tier tech firm, are we a tech giant or a call center?”

The Logic: Numbers don’t lie. If your RPE is under $150k, you are a service business. Own it, or fix it.

The Verdict:

• 0-3 “Yes” Answers: You are a pure Product Company. Stay lean, stay ruthless.

• 4-7 “Yes” Answers: You are in the “Danger Zone.” You’re a product company that’s slowly being eaten by the “Service Virus.”

• 8-10 “Yes” Answers: Congratulations, you work at a body shop. Update your LinkedIn and find a company that actually values leverage.

The Closing of the Loop

Back at ZenithTech, the eighty “resources” arrived. They were lovely people, but they had no idea how our core architecture worked. Because the culture was hierarchical and “face-saving,” they never told Sanjay they were lost. They just kept quiet and moved tickets around in Jira to look busy.

The senior engineers—the ones who actually built the product—spent 90% of their time in “Onboarding Sessions” and “Code Reviews” that felt more like basic literacy classes. Our “Velocity” went negative. We were actually losing features because the new guys kept overwriting core logic.

Sanjay was ecstatic. He showed the board a chart showing “Headcount Growth” as a proxy for “Productivity Growth.”

Three months later, the European client canceled the contract. The software was too buggy to deploy. ZenithTech’s RPE plummeted, and the “Product” was now a “Service Monster” that required 400 people to keep the login screen from crashing.

The tragedy of modern tech is that we’ve forgotten what productivity looks like. It doesn’t look like a crowded office or a long Jira board. It looks like a small, quiet room where three brilliant people are saying “No” to a hundred bad ideas so they can say “Yes” to the one that changes the world.

Stop counting the shovels.

LEGAL DISCLAIMER AND SURVIVAL GUIDE: The following article is a work of high-octane, hallucinogenic fiction. If you read a sentence and think, “Wait, that sounds exactly like my Monday morning stand-up,” that is just your suppressed corporate trauma speaking. Any resemblance to actual persons (living, dead, or currently wearing a Patagonia vest in a coworking space), actual companies, or actual “game-changing” cloud telephony startups is entirely coincidental—in the same way that a power cut in Bangalore hitting exactly when you have a critical deployment is a “coincidence.” The events described have been exaggerated to a degree that would make a Bollywood director blush, specifically to drive the point home without needing a 400-slide deck. In absolute reality, the CEO of the actual company was a legitimate gem of a human being, a strategic wizard who built a top-tier team and successfully navigated us to a glorious acquisition while the rest of us were still trying to figure out how to work the office printer. No CEOs, engineers, or coconuts were harmed in the making of this narrative. Mostly.

The conference room in our MG Road office in Bangalore felt like a pressurized cabin in the middle of a monsoon. It was second half of 2015, and our cloud telephony startup was supposedly “disrupting” how India talked. Outside, the MG Road traffic was a gridlocked symphony of frustrated horns, neon signs, and the smell of wet asphalt. Inside, our CEO—a man who once spent twenty minutes explaining how a coconut was a metaphor for a scalable backend—stood at the head of a mahogany table. He didn’t look at the catastrophic churn numbers on the screen. He didn’t look at the lead engineer, who was vibrating with a rage that only a week of failed SIP trunk deployments can produce.

Instead, he pointed to a slide that featured a single, lonely circle in the middle of a vast white space. “This,” he whispered, “is the void. And our five-year plan is the light that will fill it. We aren’t just routing calls for e-commerce delivery boys anymore. We are architecting the collective consciousness of the mobile-first consumer.”

I looked at my notepad. My only note from the last hour was a doodle of a tombstone with our company logo on it. We were a telephony platform. Our “light” was currently flickering because our primary API was held together by duct tape and prayers. But in that room, the “Void” was more important than the fact that our biggest taxi-aggregator client was currently on the phone with our lawyers because their drivers couldn’t reach their customers.

This is the central pathology of tech. We are so in love with the Light that we forget how to walk in the dark. We treat strategy like a religious text, planning like a prophecy, and execution like a chore for the help.

1. Strategy is the Art of Brutal Subtraction

Strategy is the most abused word in the corporate dictionary. Most people think strategy is a thirty-page slide deck filled with words like “synergy” and “leveraging.” It isn’t. Strategy is simply the art of deciding what you are not going to do.

The Case of the Chinese Wall: The Trade Desk vs. Google

In the late 2000s, the adtech world was a wild west of “do-it-all” giants. Google was building a “full-stack” solution. They wanted the supply side (the publishers), the demand side (the advertisers), and the data marketplace in the middle. They wanted to be the auctioneer, the buyer, the seller, and the guy who owned the building where the auction happened.

Jeff Green, the founder of The Trade Desk, saw this and made a choice that looked like suicide at the time. He focused only on the demand side (the buyers). For the first ten years—long after becoming a public and NASDAQ 100 company—he refused to touch the supply side.

Why? Because it created a strategic moat of trust. Meanwhile, Google was playing a dangerous game. By owning both sides, they created a massive conflict of interest. This lack of a “Chinese Wall” eventually led to the infamous Project Bernanke. While it only came to light years later through legal filings, the groundwork was laid in the late 2000s. Google allegedly used its “God’s eye view” of the market to manipulate auctions, taking data from other advertisers to give their own bids a secret advantage. They weren’t just the house; they were playing at the table with marked cards.

Google’s strategy was “total dominance,” which is great until the market realizes you’re playing a rigged game. The Trade Desk’s strategy was total alignment. They chose not to take the easy money from publishers so they could be the undisputed champion for advertisers. That is a strategy. If your strategy doesn’t hurt a little bit, it isn’t a strategy; it’s just a hobby.

“I’m as proud of many of the things we haven’t done as the things we have done. Innovation is saying no to a thousand things.” — Steve Jobs

2. The North Star: Navigation vs. Decoration

Every startup in Bangalore in 2015 had a “North Star Metric.” For some, it was “Total Registered Users.” For others, it was “Number of Cities Launched.” Ours was “Minutes of Usage.”

The North Star is supposed to be the one high-level metric that best captures the core value that your product delivers. If that number goes up, the company is winning. If it goes down, you are dying.

The Vanity Metric Trap

Most companies pick a North Star that looks good on a pitch deck but means nothing for the bottom line. If your North Star is “Registered Users,” but only 2% of them ever make a second phone call, your North Star is actually a black hole.

A real North Star for a cloud telephony company should have been “Successful Call Completions.” That is the only thing the customer actually pays for. But that was a hard number to move because it required fixing the messy reality of Indian telecom infrastructure. It was much easier to track “App Downloads.”

• Airbnb: Their North Star isn’t “App Installs.” It is “Nights Booked.” That aligns the host, the guest, and the company.

• Practicality Check: If your North Star hits a million but your bank account is empty, did you actually succeed? If the answer is no, throw that metric in the bin.

3. Goals: The Three-Year Fairytale

In tech, we love to pretend we are architects, but we are actually more like gardeners. An architect draws a blueprint and the building doesn’t move. A gardener plants a seed, and then the weather, the bugs, and the soil decide what actually happens.

Theoretical planning tells you to build a three-year roadmap with quarterly milestones. Practicality tells you that a three-year roadmap in tech is just fan fiction for your board of directors. You need to break your goals into three distinct horizons:

• Short-Term (The Dirt): This is the next two weeks. This is where you track every line of code. If you don’t know exactly what is happening on Monday at 9:00 AM, you aren’t planning; you are dreaming.

• Mid-Term (The Horizon): This is the next three to six months. These are bets, not promises. You might say: “We want to reduce latency by 20%.” You don’t know exactly how yet, but that is the target.

• Long-Term (The Star): This is the next year and beyond. This is just a direction. It should be broad enough to allow for a total pivot if the market explodes.

Jeff Bezos once said, “We are stubborn on vision. We are flexible on details.” Most tech managers do the opposite. They change their vision every time a competitor releases a feature, but they are stubborn about the stupid details of a project plan made six months ago.

4. KPIs: The Math of Reality

If the North Star is the destination and Goals are the milestones, KPIs (Key Performance Indicators) are the gauges on your dashboard. They tell you if you have enough fuel or if the engine is overheating.

The Trinity of Metrics

“In business, the idea of measuring what you are doing, picking the measurements that count like customer satisfaction and performance, you thrive on that.” — Bill Gates

To stay honest, you need three specific types of data:

1. Leading Indicators: These tell you what is going to happen. For us, it was “New API Keys Generated.” More keys today meant more call volume in three months.

2. Lagging Indicators: These tell you what already happened. Revenue and Churn are lagging. By the time they look bad, the damage was done months ago.

3. Health Metrics: These are the constraints. For us, it was “Uptime.” You can hit revenue goals, but if your uptime is 90%, you are a dead man walking.

5. Tactical vs. Strategic: The Build-Buy Suicide Note

“Value is discovered by what goes into the basket, not by what is promised on the label.” — Thomas Edison

This is where the bodies are buried. A tactical decision is a shortcut to hit a deadline. A strategic decision is an investment in your core identity.

In our Bangalore office, we needed a billing engine. We could build it in-house (9 months) or “buy” it via a third-party API (2 weeks). Tactically, buying was the “smart” move. We hit our targets, and the CEO got to tweet about “velocity.”

But two years later, that vendor was bought by a direct competitor. Suddenly, our “strategic engine” was owned by the enemy. They saw our traffic patterns and raised our prices until our margins vanished.

Case Study: Apple vs. Google Maps

Early iPhones shipped with Google Maps. Tactical win? Yes. Strategic disaster? Absolutely. They gave their primary rival (Google) mountains of location data. Apple eventually spent billions to build Apple Maps from scratch to regain their sovereignty.

6. The Competition: The Rear-View Mirror Trap

Competitive research is like looking in the rear-view mirror while driving a rickshaw at 100 km/h. It’s good to know where the other cars are, but if you stare too long, you’re going to hit a cow.

In 2015, we were obsessed with Twilio. Every meeting was: “What is Twilio doing?” We were so busy trying to build a Twilio-killer that we ignored our own customers who were screaming for better reliability in rural India. We were building APIs for developers in San Francisco while our users in Mysore couldn’t get a dial tone.

When to Ignore, When to Attack

The giants in adtech were afraid to be transparent because their margins relied on a “Black Box.” The winners weren’t those who copied features; they were the ones who out-trusted the competition.

If you spend all your time looking at the other guy, you end up making a “me-too” product. Use the competition only to see where they are failing their customers. Use their mistakes as your roadmap. Don’t copy their features; copy their solutions to problems they haven’t solved yet.

“The competitor to be feared is one who never bothers about you at all, but goes on making his own business better all the time.” — Henry Ford

7. Execution: The Religion of the “Friday Pulse”

“Ideas are easy, Execution is everything.” — John Doer

Execution is the sound of a thousand small gears turning in the same direction. To drive it, you must kill the “Status Update” meeting—the one where everyone says “on track” while the Jira board looks like a crime scene.

High-Performance Discipline

• The Weekly Detail: If a task takes longer than a week, it’s a project. Break it down. “Working on the database” is a lie. “Migrated user table to new schema” is execution.

• The Friday Pulse: Every Friday, ask: “Did the SIP trunk failover test pass? Yes or no?” No excuses.

• Tolerance for Mistakes: Celebrate the failures that teach (e.g., a new architecture that didn’t scale). Fire the people who make the same mistake three times because they didn’t check the logs.

• Alignment: Driving alignment across silos means telling the sales team “No” and the engineers “Not now.” You build alignment by making everyone’s bonus dependent on the same goal.

8. The Trinity: Strategy, Planning, and Execution

To win, you must treat these as a circular ecosystem, not a linear line.

The Bridge Process

1. Strategy (The Why): The choice of where to play and where to ignore. This stays stable for years.

2. Planning (The How): The bridge between vision and work. Updated quarterly. Sets the KPIs.

3. Execution (The Do): The daily grind. The Friday Pulse. This is the only part that is real.

9. The Friday Blood Sync: The Cross-Functional War Room Checklist

I. The Binary Commitment Wall (No “90% Done” Allowed)

Every department lead has 60 seconds to report on their one major “Brick” from last week.

• Engineering: “Did the latency-reduction patch for the Singapore gateway go live?” [Yes/No]

• Product: “Is the final PRD for the ‘Call-Masking’ feature locked and approved by Legal?” [Yes/No]

• Sales: “Did we close the high-volume pilot with the Delhi logistics giant?” [Yes/No]

• Customer Success (CS): “Have all Tier-1 tickets regarding the ‘Dropped Call’ bug from Tuesday been resolved?” [Yes/No]

• Infrastructure/DevOps: “Was the database migration completed without a packet loss event?” [Yes/No]

The Rule of Shame: If the answer is “No,” the lead must explain the Blocker, not the Excuse. Was it a dependency on another team? A tactical mistake? Or did we just overestimate our own brilliance?

II. The Cross-Functional Friction Audit

This is where we find out who is accidentally sabotaging whom.

• The Sales vs. Engineering Trap: Did Sales promise a “custom integration” to a client this week that isn’t on the roadmap? If so, kill the promise or kill the roadmap. You cannot have both.

• The Product vs. CS Gap: CS reports the top 3 complaints from the week. If Product isn’t working on a fix for at least one of them, justify why we are ignoring our customers.

• The Marketing vs. Reality Check: Is Marketing promoting “99.99% Uptime” while Infra is currently holding the servers together with prayers and caffeine? Alignment on the “Truth” is mandatory.

III. The Dashboard of Hard Truths (Hard Metrics)

No “vibe-based” metrics. If it isn’t a hard number, it’s a hallucination.

IV. The Strategic Veto: The Art of Saying “No”

To maintain ruthless prioritization, we must identify what we are dropping to stay fast.

4. The “Shiny Object” Cull: What was the most distracting idea suggested this week? (e.g., “Let’s add AI to the dialer!”) Action: Officially veto it for the next 90 days.

5. Technical Debt Payment: What feature did we decide not to build this week so that Engineering could fix a core stability issue?

V. The Monday Morning “Zero-Hour” Plan

Execution isn’t finished until the pivot for next week is set.

• The Carry-Over Tax: Any “No” from Section I is automatically the Top Priority for Monday 9:00 AM.

• The Clean Slate: If a team is consistently failing their binary commitment, do we need to reduce their scope or change the lead? (Be brutal, be honest).

• The “Delhi” Sign-off: Before the CEO catches that flight back to Delhi, does he agree that this week’s “Bricks” actually support the “Bridge”?

The Closing Question: “Is there any silent disaster currently brewing that no one has mentioned yet?”

(Silence for 10 seconds. If someone speaks up, stay another hour. If not, go home.)

10. Prep To Prevent The Friday Blood Sync: Get The Strategy and Plan Right

To ensure the “Friday Blood Sync” doesn’t turn into a finger-pointing exercise, you need the underlying scaffolding to be just as ruthless. If the Execution is the engine, the Strategic and Planning frameworks are the chassis and the GPS.

Here is the three-section breakdown for the Strategic Framework, the Planning Framework, and the final Measurement Scorecard.

I. The Strategic Framework: The “Moat & Anchor” Model

This framework is used only once a quarter (or whenever a massive market shift occurs). Its goal is to define the boundaries of your universe.

1. The Strategic Filter (The “Moat”)

Every potential initiative must pass through the Moat Test. In our cloud telephony world, our moat was “Unrivaled Latency in Tier-2 Indian Cities.”

• The Question: Does this project deepen the moat?

• The Outcome: If it doesn’t improve call quality or reliability, it is a “Secondary Priority.” No matter how “cool” the feature is.

2. The Subtraction List (The “Anchor”)

Strategy is defined by what you drop.

• The List: Formally document the three high-revenue opportunities you are ignoring this quarter to stay focused.

• Example: “We are ignoring the Enterprise Video market to own the SMB Voice market.”

3. The Chinese Wall Protocol (Integrity)

Referencing our Trade Desk vs. Google lesson:

6. The Rule: Define where your interests stop to maintain trust.

7. Action: Document the “No-Go” zones. (e.g., “We route the calls; we never compete with our customers by launching our own call center.”)

II. The Planning Framework: The “Rolling Horizon” (70-20-10)

Planning is not a fixed document; it is a living breathing organism. We use the 70-20-10 rule to ensure we aren’t hallucinating about 2029 while the 2026 servers are smoking.

1. The 70% (The Dirt - 2-Week Sprints)

• Granularity: High. Individual tasks, owners, and hourly estimates.

• Flexibility: Zero. Once a sprint starts, the only way to change it is to declare a “State of Emergency.”

2. The 20% (The Horizon - 3-Month OKRs)

4. Granularity: Medium. These are “Key Results” (e.g., “Migrate 40% of traffic to the new SIP gateway”).

5. Flexibility: Moderate. If the “Dirt” reveals a major technical blocker, we pivot the Horizon.

3. The 10% (The Star - 1-Year Vision)

• Granularity: Low. Themes and North Stars.

• Flexibility: High. This is our “Fan Fiction.” It exists to keep the CEO’s “Delhi” dreams aligned with the Bangalore reality.

III. The Measurement Scorecard: The “Hard Truth” Ledger

This is the final scorecard that combines the Strategy, Planning, and Execution into one brutal document. It is the only thing that matters during the Friday Blood Sync.

IV. The “Perfect Execution” Audit

To ensure this isn’t just more “theoretical practice,” the leadership must answer three questions every Sunday night before the week begins:

1. The Sovereignty Check: Are we “Building” the core and “Buying” the fluff? (Did we accidentally outsource our soul to a vendor this week?)

2. The Competition Blindfold: Are we building this feature because a competitor released it yesterday or does it really help us differentiate in the market to add value to our customers’ business ?

3. The Alignment Sanity: If I ask a junior developer and the Sales Lead what our #1 goal is, will they give me the same answer?

11. The Closing: Back To The Story

The boardroom in MG Road finally emptied around 4:00 AM. The “Void” slide was still glowing, but the CEO had already caught the red-eye flight back to Delhi. I stayed behind with the lead engineer. We didn’t talk about “sentimental resonance engines.”

We sat in the dark, listening to the rain, and talked about rewriting three lines of code so delivery drivers could receive calls at breakfast. We looked at the real KPIs—the ones showing our server was about to melt—and ignored the “North Star” of App Downloads.

That company did get acquired after a few years. Everyone did make some money. But they were so busy being “strategic” that they forgot to be functional. They had the map and the route and they could have made it much bigger than Twilio, but they let the engine catch fire because they were too busy arguing about the destination.

Hindsight is a great teacher.

Strategy is your map. Planning is your route. Execution is your engine. If you spend all your time looking at the map while the engine is on fire, you are never going to get where you are going.

Why fixing your font size won’t save you from the Algorithm, and why you need a GPS, not a prayer. Try CareerPlot if you haven’t.

Part I: The Funeral of a Job Title

Let’s start with a scene you probably know too well. Let’s talk about Jason.

Jason is a Product Lead at a mid-sized tech company. For six months, Jason has been feeling uneasy. But he pushes it down with the help of too much coffee, a false sense of security, and the comforting glow of his dual monitors.

Last Tuesday, at 9:00 AM sharp, Jason joined a mandatory all-hands video call. The CEO was calling in from what looked suspiciously like a ski lodge in Aspen, wearing a turtleneck that cost more than Jason’s car. He spoke somberly about “macroeconomic headwinds” and “right-sizing the ship to optimize operational synergy.” Jason zoned out because he was too busy messaging his work bestie, Sarah, making a joke about how the CEO looked like a Bond villain who just bought a Peloton.

By 9:04 AM, Jason’s screen went black. His Slack access was gone. His email was locked. His job title evaporated into thin air. He was part of the 12% workforce reduction.

By 9:30 AM, Jason was panicking. By 10:00 AM, he was in the Denial Phase. And by 10:15 AM, he was doing what every desperate professional does when their career implodes. He opened a Word document.

For the last three weekends, Jason has treated his resume like it is a sacred text. He has polished it like a diamond. He agonized over whether “spearheaded” sounds more authoritative than “led.” He sprinkled in corporate words like “cross-functional collaboration” and “dynamic innovator” as if they were magic spells meant to exorcise the demon of unemployment.

He wrote a stunning piece of corporate fan fiction starring himself as the hero. He hit the Easy Apply button on two hundred jobs on LinkedIn. Then he sat back, staring at his inbox, waiting for the gods of recruiting to descend from the clouds and crown him the Chosen One.

And what did he get?

Deafening silence. Broken only by the occasional automated rejection email from a robot named “No-Reply” telling him they decided to move forward with other candidates.

Here is the cold, brutal truth that nobody wants to admit. Your resume is just a blood test.

It is a boring, clinical readout of data points. It confirms you were, in fact, present at a desk from 2021 to 2023. It proves you were alive. It does not explain that you spent half that time staring into the abyss of your lukewarm coffee questioning every life choice that led you to a Tuesday morning meeting about TPS reports.

As Warren Buffett famously said, “Only when the tide goes out do you discover who’s been swimming naked.”

Well, Jason. The tide is out. And you are naked.

Part II: The Market Has No Morality (And It Hates Your Rent)

We need to be honest about the dumpster fire that is the current job market. For the last ten years, we lived in a fantasy land.

Money was free. Interest rates were zero. Tech companies hired people just so their competitors couldn’t hire them. You could get a job as a “Happiness Manager” for a hundred and forty grand a year just by vibing correctly in the interview.

That party is over. The lights are on. The music stopped. And the police are here.

According to Layoffs.fyi, over 260,000 tech workers lost their jobs in 2023, and another 140,000+ followed in 2024. This isn’t a “blip.” It is a correction.

Companies are no longer optimizing for growth. They are optimizing for efficiency. When ten thousand highly qualified engineers from Google and Meta flood the market on the same Tuesday, your proficiency in Microsoft Excel suddenly looks a lot less impressive.

You are no longer competing with the lazy guy who sleeps under his desk. You are competing with the guy who built the desk, the software running on the desk, and the AI that is currently rewriting your cover letter.

As Jack Welch, the former CEO of GE, ruthlessly put it: “Face reality as it is, not as it was or as you wish it to be.”

The reality is that “spray and pray” applications are dead. Data from recruiting platforms shows that applicants on LinkedIn have a response rate of just 3% to 13%. Compare that to employee referrals, which are 4x more likely to result in a hire.

Applying to fifty jobs a day without a strategy isn’t grit. It’s spam. You are just buying lottery tickets with your time.

Part III: AI Is Not a Skill, It’s Oxygen

Here is the part where Jason thinks he is safe because he knows how to use ChatGPT to write a limerick.

Jason lists “Generative AI” as a skill on his resume, right next to “Microsoft PowerPoint.” He thinks this makes him special.

Jason is wrong.

AI is going to complicate the job scenario in ways that most people are not ready for. A recent Goldman Sachs report estimates that AI could displace 300 million full-time jobs globally and that 25% of all work tasks in the US could be automated.

Here is the kicker: Junior roles are disappearing. A study on the impact of ChatGPT showed that job postings for junior software developers dropped by 20% compared to senior roles. Why? Because AI can do the junior work. It can write the basic code. It can draft the basic email. It can summarize the meeting.

If your job is “taking information from Pile A and moving it to Pile B,” you are in trouble.

Being able to “use” AI is no longer a competitive advantage. It is table stakes. It is like knowing how to type. You don’t get a medal for knowing how to use a keyboard, and you won’t get a job just because you can write a prompt.

The bar has been raised. You need to be the person who directs the AI, not the person who just chats with it. You need high-level strategic thinking because the low-level execution is now free.

As Jensen Huang, the CEO of Nvidia, said: “It is not AI that will take your job, but the person who uses AI that will take your job.”

If you are not using these tools to become 10x faster, you are already behind the curve.

Part IV: The Frog in Boiling Water (Ignoring the Signals)

The most painful part of this story isn’t the layoff itself. It is the shock. It’s the sheer, unadulterated surprise on Jason’s face when his access card stops working.

Most people navigate their careers with the situational awareness of a toddler wandering through a construction site. They believe that if they just “do a good job,” they are safe. This is adorable. It is also dead wrong.

We are currently living through a corporate “Vibe Shift” of catastrophic proportions, and you-like Jason-are probably ignoring the flashing red lights because they are inconvenient.

The “Quiet Firing” Playbook

Let’s look at the data. According to recent workforce reports, nearly 70% of companies admit to using “Quiet Firing” tactics-subtle changes designed to make you miserable enough to quit so they don’t have to pay you severance.

Jason missed the signals because he thought they were just “policy changes.” They weren’t. They were smoke signals.

Signal #1: The Snack Downgrade Index

It started three months ago. The office kitchen used to be stocked with Kind Bars, coconut water, and artisanal beef jerky. Then, one Monday, Jason walked in to find… pretzels. Just a massive, sad tub of pretzels and a coffee machine that had been downgraded to “generic sludge.”

Signal #2: The “Return to Office” Mandate (With a Twist)

Suddenly, the CEO who spent two years preaching about “remote-first flexibility” sent a memo titled “Re-igniting Our Culture.” It demanded everyone be in the office 4 days a week.

Signal #3: The “Ghost” Calendar

Remember when Jason’s boss, “Chad,” used to micromanage him? Chad was annoying. But then, Chad stopped. He cancelled their weekly 1:1 three weeks in a row. He stopped asking for status updates.

Signal #4: The Consultant Invasion

Two months ago, a group of people in sharp suits from “Strategy Corp” started wandering the halls. They didn’t talk to Jason. They just asked for “access to the data room” and “org chart visualization.”

We prefer the comfort of routine to the discomfort of reality. We tell ourselves lies like, “They can’t fire me, I’m the only one who knows how to run the legacy reporting system.”

Newsflash: They don’t care about the legacy reporting system. They will replace it (and you) with a Python script and a twenty-two-year-old intern named Kyle who runs on Red Bull and desperation. And they will do it in a weekend.

Jason was comfortable. He thought his seniority was a shield. He forgot the golden rule of business survival, perfectly summarized by Bill Gates:

“Success is a lousy teacher. It seduces smart people into thinking they can’t lose.”

Jason thought he couldn’t lose because he hadn’t lost yet. He sat in the pot, feeling the water get warmer, telling himself it was just a nice jacuzzi, right up until the moment he was boiled alive.

Part V: Maybe You Are Climbing the Wrong Tree

Here is a question that hurts even more than the layoff. What if you shouldn’t even be applying for these jobs?

We get so obsessed with getting a job that we forget to ask if we want this job.

According to Gallup’s State of the Global Workplace report, only 23% of employees are actually engaged at work. That means nearly 80% of people are just showing up, doing the bare minimum, and collecting a check.

I see so many people desperate to get back into roles they hated. They were miserable as Product Managers. They hated the politics. They hated the stakeholders. But the moment they lose the job, they are desperate to get it back. Why? Because of the sunk cost fallacy.

Maybe you aren’t a bad employee. Maybe you are just a fish trying to climb a tree.

You might have excellent skills in analysis, but you are forcing yourself into sales because that is where the commission is. You are setting yourself up to be mediocre and miserable.

The market is brutal right now. If you are only 70% committed to a role, you will be crushed by the person who is 100% committed. You cannot fake enthusiasm when you are competing against fanatics.

Steve Jobs said it clearly: “Your work is going to fill a large part of your life, and the only way to be truly satisfied is to do what you believe is great work. And the only way to do great work is to love what you do.”

If you are just looking for a paycheck, the person who loves the work will beat you every single time.

Part VI: The “Open To Work” Desperation Banner

Now, let’s talk about the green ring of death on LinkedIn. The #OpenToWork banner.

I know, I know. It is supposed to be a community signal. “Help me!” But let’s be brutally honest about human psychology and negotiation leverage.

When you slap that green frame on your photo, you are essentially walking into a bar, standing on a table, and screaming, “I am currently single and very, very lonely! Please, someone buy me a drink!”

Does that make you attractive? No. It makes you look desperate.

Recruiters are like cats. If you chase them, they run away. If you ignore them, they jump in your lap.

When a recruiter sees “Open to Work,” they subconsciously register a few things:

1. This person has no current leverage.

2. I can lowball their salary offer.

3. Why hasn’t anyone else snapped them up yet? Is there something wrong with the merchandise?

It is a harsh, unfair psychological bias. But it is real. The best candidates-the ones who get headhunted-are the ones who look like they are too busy being successful to care about a LinkedIn banner. They project value, not availability.

As Charlie Munger, the late partner of Warren Buffett, wisely said: “It is remarkable how much long-term advantage people like us have gotten by trying to be consistently not stupid, instead of trying to be very intelligent.”

Broadcasting your desperation to the world is not smart. It destroys your leverage before you even enter the room.

Part VII: Structure is the Antidote to Chaos

So if resumes are a distraction, the job market is rigged, AI is coming for your lunch, and your instincts are wrong… what’s left?

Structure.

I have been a Product Manager for a long time. In product, we don’t just build things and hope they work. We have a roadmap. We have data. We have a plan.

You need to treat your career like a product.

Peter Drucker, the father of management thinking, said: “What gets measured gets managed.”

Most people do not measure their job hunt. They operate on vibes and panic. They wake up and ask, “What should I do today?”

That is the wrong question. You should already know what you are doing today because you have a system.

This is where CareerPlot comes in. It isn’t a magic wand. It is a set of guardrails for the chaos.

We need to move away from the “guru” model where a billionaire tells you to “crush it,” and move toward a structural model where you actually have a plan.

1. The Reality Check (AI Resume Analyzer)

First, you need the truth. CareerPlot uses an AI that acts like a ruthless hiring manager. It doesn’t just check your grammar. It runs a SWOT analysis (Strengths, Weaknesses, Opportunities, Threats). It might tell you that your industry is shrinking by 5% annually, or that your skills gap is the reason you aren’t getting callbacks. It hurts, but it’s the data you need.

2. The Whole Life Audit (AI Coach)

Most career advice fails because it ignores your life. CareerPlot’s AI Coach runs an 18-Dimension Life Assessment. If you have high financial obligations and low risk tolerance, it won’t suggest you join a volatile crypto startup. It aligns your career path with your actual reality.

3. The Roadmap (Career Tracker)

The Career Tracker kills the “spray and pray” method. It gives you a weekly action plan. “Learn this specific AI tool.” “Message these three alumni.” “Fix this portfolio item.” It gamifies the process so you stop doom-scrolling and start executing.

Part VIII: Closing the Tab

Let’s go back to Jason.

It’s now 4:00 PM. The sun is setting on his first day of unemployment. He has twelve versions of his resume saved on his desktop. Jason_Resume_FINAL, Jason_Resume_FINAL_v2, Jason_Resume_REAL_FINAL.

He realizes, finally, that version 13 isn’t going to change anything.

He closes the Word document. He closes the 47 tabs of “Easy Apply” jobs he didn’t actually read. He takes down the “Open to Work” banner because he realizes his value isn’t defined by his availability.

He opens CareerPlot. He stops treating his career like a lottery ticket and starts treating it like a project.

He runs the analysis and realizes his lack of AI-tooling experience is a major threat. He sets up the tracker and assigns himself a task: “Learn Prompt Engineering Basics — Due Friday.” He stops panicking about the “market” and starts building his own market value.

For the first time all day, the panic subsides. The chest tightness loosens. He has a plan. He isn’t just hoping for a job anymore. He is engineering his next move.

Your career isn’t collapsing from a lack of ambition or a shortage of dreams. It’s collapsing because you are drowning in useless advice, starving for a simple plan, and there’s nobody around to call you on your own excuses.

Stop polishing the blood test. Start working on the patient.

Ready to stop guessing? Build your structure at CareerPlot.com.

Originally published at https://www.amitgoel.me on January 30, 2026.

Congratulations on Your Perfect Resume. Now Let’s Talk About Your Actual Career. was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

Lock your doors, hide your high-yield coffee beans, and delete your search history for “how to explain a 4% raise during 8% inflation.” It’s that magical time of year again: The Annual Appraisal Cycle.

Forget The Last of Us or Squid Game; those are rom-coms compared to the psychological warfare of the Corporate Finance Bell Curve. This is the time when your manager-who hasn’t seen a spreadsheet cell since the Obama administration-suddenly becomes a mathematical savant, hell-bent on proving that despite your 80-hour work weeks and “10x impact,” you are, statistically speaking, just “aggressively average.”

Welcome to the Corporate Hunger Games, where the Tribute from Accounting must fight the Tribute from Compliance over a single “Exceeds Expectations” slot that was actually promised to the CEO’s nephew three months ago.

Imagine you’ve just spent the last twelve months absolutely crushing it. You shipped the core API three weeks early, you mentored four juniors until they stopped crying in the breakroom, and you saved the company’s flagship product from a catastrophic security breach while the rest of the leadership team was at a “mindfulness retreat” in Bali.

You walk into your annual review feeling like a gladiator returning from a successful conquest. You’re ready for that “Exceeds Expectations” rating, a bonus that makes your mortgage look small, and a promotion that actually reflects your impact.

But as the door closes, the atmosphere shifts. Your manager, Dave-who has the spinal column of a chocolate eclair-won’t look you in the eye. He starts talking about “organizational health,” “budgetary constraints,” and “the broader talent distribution.”

Translation: The HR department just ran your career through a 200-year-old math equation, and you lost.

“Look,” Dave stammers, “You did an incredible job. Truly. But we already gave the ‘Exceeds’ rating to Sarah because she’s been here longer, and the system says only 10% of the team can be ‘Great.’ So, you’re officially ‘Solidly Meets Expectations.’ Here’s a 2% raise and a branded hoodie. Don’t spend it all at once.”

This is the Bell Curve. It is a mathematical mugging. It is the professional equivalent of telling a marathon runner they finished in “last place” because the race director decided that only three people are allowed to be fast today. It is the single most effective way to turn your best employees into your competitors’ best employees by next Monday.

The Gaussian Ghost: A Brief History of Scientific Laziness

To understand why we are still stuck in this nightmare, we have to look at where it came from. The bell curve, or Normal Distribution, was popularized in the early 19th century by Carl Friedrich Gauss. He didn’t design it to measure software engineers or product leads; he used it to track errors in astronomical observations and the physical heights of French conscripts.

It was a way to measure accidents, errors, and static physical traits. It was never meant to measure human potential, creativity, or the non-linear impact of a knowledge worker.

Then came the 1980s. Jack Welch, the then-CEO of General Electric, decided that the best way to run a company was to treat it like a gladiatorial pit. He popularized “Rank and Yank” (also known as the “Vitality Curve”). You put 20% of your people in the top tier, 70% in the middle, and 10% at the bottom. Then, you fired the bottom 10%.

It worked for GE back then because they were an industrial conglomerate making lightbulbs and jet engines. In a factory setting, the difference between a “good” worker and a “bad” worker is marginal. If a guy on the assembly line is 10% faster, he’s a star.

But in the 21st century, we don’t manufacture lightbulbs; we manufacture ideas. And ideas do not follow a bell curve. By forcing a normal distribution on a high-performance team, you are effectively telling your overachievers that their hard work is statistically inconvenient.

The Myth and the Monster: The 10x Engineer

Whenever you criticize the bell curve, an HR “Thought Leader” (usually someone who has never written a line of code or closed a million-dollar deal) will pipe up: “But the 10x engineer is a myth! It’s all about the team!”

Let’s be brutally honest: People who say 10x engineers don’t exist are usually 1x engineers (or 0.5x “B-players”) who are terrified of being measured against them.

What is a 10x Engineer?

A 10x engineer isn’t someone who types ten times faster. They aren’t the ones staying until 2 AM every night to show how “busy” they are. In fact, a 10x engineer often looks like they’re doing less work.

The 1x engineer solves a problem by writing 2,000 lines of complex, brittle code. The 10x engineer spends three hours staring at a whiteboard, realizes the problem shouldn’t exist in the first place, deletes 500 lines of old code, and replaces the whole mess with a 10-line elegant script.

How to Identify Them:

• The Simplifiers: While everyone else is adding features, the 10x-er is removing friction.

• The Force Multipliers: They don’t just do their work; they build tools and systems that make everyone else on the team 2x better.

• The Outcome Obsessed: They don’t care about “Jira tickets closed.” They care about whether the user’s problem was actually solved.

In a bell curve system, the 10x engineer is an anomaly that HR tries to “average out.” But in reality, performance in the tech world follows a Power Law (Pareto Distribution). A tiny fraction of the people produces the vast majority of the value. Trying to put a 10x engineer on a bell curve is like trying to fit a skyscraper into a shoebox.

The Anatomy of the “Brilliant Jerk” (Netflix’s Warning)

The term “Brilliant Jerk” was Netflix’s gift to the world. A Brilliant Jerk is that person who is technically untouchable-the architect who can recite the entire kernel source code from memory-but who treats everyone else like cognitive peasants. They use their “brilliance” as a shield to bully colleagues, ignore processes, and create a “fear-based” environment.

Why the Bell Curve is Their Natural Habitat:
In a forced-ranking system, the Brilliant Jerk is safe. Why? Because the system only measures “Output.” The Jerk produces high output, so they get the “Exceeds” rating. Meanwhile, the five people they bullied into silence or sabotaged are placed in the “Average” bucket because their “productivity dropped.”
Netflix realized that the “Brilliance” of one jerk is never worth the collective productivity loss of the ten people they’ve demotivated. If you have a Brilliant Jerk, you don’t have a high-performer; you have a human dumpster fire with a high IQ. They should be fired immediately, regardless of how “critical” they claim to be.

The Hero Syndrome: A Saboteur with a Cape

We’ve all met the “Hero.” This is the person who swoops in at 3 AM to fix a crashing server or the “Firefighter” who saves a failing account at the last second. The company gives them a plaque. The CEO mentions them in the All-Hands.

Here is the brutal truth: If that person has saved the company more than twice, they aren’t a hero-they are a liability.

Hero Syndrome is often a Brilliant Jerk in disguise. These individuals thrive on crisis. They build “black box” systems that only they understand. They don’t document their work. They subconsciously (or consciously) ensure that the company is constantly on fire so they can be the only one with a hose.

The Test: If your “Hero” goes on vacation and the company falls apart, they haven’t saved you; they’ve sabotaged you. A true 10x performer builds systems that don’t break and mentors others so they are never the single point of failure. If you have a “Hero” who has saved the day three times in a row, don’t give them a bonus. Fire them. They are preventing your company from becoming a professional organization.

The High Cost of the “Average” 80%

HR heads love the Bell Curve because it makes the budget predictable. “If we keep 80% of people in the ‘Average’ bucket, our salary growth remains flat. We’re saving money!”

Congratulations, you just spent $10 million to save $10,000.

A larger team of mediocre people is exponentially more expensive than a small team of superstars.

1. Management Bloat: Mediocre people need managers. Managers need directors. Directors need VPs. Soon, you have six layers of people whose only job is to “coordinate” the lack of initiative.

2. The Drag Effect: Mediocre people build mediocre products. They don’t innovate; they maintain.

3. The Recruitment Death Spiral: Superstars want to work with other superstars. The moment a 10x engineer realizes they are being “curved” alongside a C-player, they leave. You are left with a company full of “average” people who are just happy to have a job.

The Hall of Fame: The Curve Killers

These are the companies that realized that humans aren’t data points and decided to treat talent like a scarce resource.

• Microsoft (The Satya Pivot): Under Steve Ballmer, Microsoft was the poster child for “Stack Ranking.” It turned the company into a circular firing squad. Engineers were more focused on sabotaging their deskmates than beating Apple. When Satya Nadella took over, he killed the curve. He replaced “competition” with “Growth Mindset.” The result? Microsoft’s stock price entered orbit.

• Stripe: They ignore the curve entirely. They evaluate based on Operating Principles. If you don’t meet the absolute bar of being a “Stripe-level” thinker, you don’t stay. They don’t care about buckets; they care about excellence.

• Netflix: The “Keeper Test.” No formal reviews. Just one question: “If this person wanted to leave, would I fight to keep them?” If not, they get a massive severance package. It’s the ultimate removal of mediocrity.

• Google & Meta: They use Absolute Evaluation. You are measured against the Impact you had on the business. If an entire team of 50 people has a “Redefining” year, Google pays all 50 of them like gods. They understand that talent is a Power Law, not a bell.

The Hall of Shame: The Jurassic Relics

• The “Big Four” Consulting Firms (Deloitte, KPMG, PwC, EY): They love the curve because they sell “hours.” They need a massive, replaceable middle of junior associates to bill clients. It’s not about performance; it’s about survival of the most caffeinated.

• Amazon (The Ghost of the Curve): Despite their PR, the culture of “Unregretted Attrition” is legendary. Managers are often pressured to identify a bottom percentage to “move on” every year. It’s The Hunger Games with Prime shipping.

• Legacy Manufacturing (Ford, GE): Still trying to manage creative white-collar workers using the same metrics they use for stamping out truck bumpers.

The Solution: Absolute Evaluation

If you want to kill the bell curve, you must move to Absolute Evaluation Against an Infinite Ceiling.

1. Set an Absolute Bar: Define exactly what “Excellent” looks like for every role. Not “better than Steve,” but “The code is clean, the documentation is perfect, and the stakeholders are happy.”

2. The No-Quota Rule: If every single person on the team hits that bar, every single person gets the top rating and the top bonus. Yes, the budget will be higher. But your Revenue will be 10x higher because your people aren’t spending 40% of their time wondering how to make Steve look bad.

3. Real-Time Feedback: If someone is failing, tell them on Tuesday. Don’t wait for a “Performance Review” in six months to drop a bomb on their career.

4. The Jerk Tax: Add a “Cultural Impact” metric. If a “Brilliant Jerk” has high output but low cultural scores, they get a “Needs Improvement” rating.

5. Remove the Floor: Instead of firing the bottom 10%, fire the people who don’t meet the Absolute Bar. Some years that might be 2%. Some years it might be 20%. Let the work decide, not the graph.

The Final Word

The Bell Curve is a crutch for weak leaders. It is a confession that you don’t know how to measure value, so you’ve decided to measure “relative position” instead.

Back in that review room with Dave, the air is thick with the scent of corporate cowardice. Dave finishes his spiel about “averaging out” and “normalization.” He looks at you, waiting for a nod of acceptance.

But you don’t nod. You stand up.

“Dave,” you say, your voice calm and terrifyingly clear. “You aren’t managing a team. You’re managing a graph. And a graph doesn’t build software. I do. Sarah does. And the fact that you’re willing to sacrifice my performance on the altar of a 19th-century math error tells me everything I need to know about your leadership.”

You walk out. You don’t take the Starbucks gift card. You don’t take the branded hoodie.

By the time you reach the parking lot, you’ve already sent a text to a founder who knows that a 10x engineer is worth more than a thousand “average” drones. You’ve sent a message to a company that uses the Keeper Test, not the Bell Curve.

The era of managing humans like lightbulbs is over. It’s time to stop managing by math and start leading by merit. Throw the curve in the trash. Burn the spreadsheets. Your best people-the outliers, the superstars, the legends-are already halfway out the door.

Are you going to keep the curve? Or are you going to keep your talent?

Originally published at https://www.amitgoel.me on January 13, 2026.

The Annual Hunger Games : Appraisals Are Round The Corner! Get Ready !! was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

The “Oh S**t” Moment

I was sitting in Zus Coffee in Kuala Lumpur -or maybe it was a Third Wave Coffee in Bangalore or a Blue Bottle in San Francisco; the scent of venture capital desperation smells the same everywhere-listening to a pitch. The founder, let’s call him “Prompty Paul,” was sweating through his sustainable Allbirds.

“We are building the definitive interface for legal document analysis,” Paul said, his eyes gleaming with the manic energy of a man who just raised a seed round at a $20M valuation on a napkin. “We take a PDF, you ask questions, it answers. It’s a moat. A fortress. A revolution.”

I looked at him. He looked at me. Then I looked at my phone.

Notification from OpenAI: “New Feature Alert: Upload any document, analyze instantly. Free for Plus users.”

I turned the screen to Paul. The color drained from his face faster than liquidity from a crypto exchange in 2022.

“That’s… that’s just a feature,” he stammered, his voice cracking. “We have… better prompt engineering. We have a dark mode.”

Paul didn’t have a company. Paul had a button. And that button just got “Sherlocked.”

As one anonymous VC brutally put it: “If you’re building a feature that a platform can add in a weekend, you’re not a founder. You’re an unpaid intern for Satya Nadella.”

This isn’t just about Paul. This is about the brutal, hilarious, and terrifying economic reality of 2025. Welcome to the Kill Zone.

The Macro “Vibes”: A Tale of Four Empires

Before we dance on the grave of Paul’s startup, let’s zoom out. The global economic chessboard has shifted. It’s no longer a “Global Village”; it’s “Global Fight Club,” and everyone is fighting dirty.

The United States is essentially three hedge funds and an Nvidia H100 cluster in a trench coat. The economy is a casino with nukes. While economists celebrate a “soft landing,” the reality is that the US is betting the entire farm on AI. We aren’t just spending money; we are setting it on fire. Bain & Company projects the AI market to hit nearly $1 Trillion by 2027, and JPMorgan forecasts that the “Mag 7” hyperscalers alone will burn through $400 Billion in Capex in 2026.

It’s a “Strategic Bubble.” If the bet pays off, the US owns the productivity of the next century. If it fails, the dollar collapses. As Sam Altman, CEO of OpenAI, says with terrifying optimism, “The socioeconomic value of linearly increasing intelligence is super-exponential… we see no reason for exponentially increasing investment to stop.”

Europe, meanwhile, has decided to sit this one out and become a Luxury Nursing Home. While America builds the Matrix, Europe is busy writing a 4,000-page white paper on the GDPR implications of the Matrix. The EU AI Act is fully online, creating a “compliance moat” so deep that only American giants can afford the lawyers to cross it. Arthur Mensch, the CEO of Mistral AI, nailed it when he said, “In Europe, we have more regulation than innovation. We need to wake up, or we will become a colony of American technology.”

Then there’s China. Do not sleep on China. The Western narrative that “Sanctions killed Chinese AI” is arrogant and dangerous. Nvidia can’t sell its best chips there? Fine. Huawei’s Ascend 910C is now reportedly matching Nvidia’s A100 performance.

While the US obsesses over ChatGPT writing high school essays, China is playing a completely different game: Industrial Sovereignty. Labs like DeepSeek and Alibaba (with Qwen 2.5) are quietly releasing open-source models that beat GPT-4 on coding and math, specifically designed to be lean and efficient. They aren’t building “Her”; they are building “Skynet meets Ford.” They are embedding intelligence into the manufacturing lines of BYD and Xiaomi to automate the physical world. They are playing “Hard Tech” while the West plays “Ad Tech.”

And finally, India. The country is at the most dangerous crossroads in its history. The “Bangalore Model” is burning. For 30 years, the pitch was simple: “We have cheaper humans than you.” Wipro, Infosys, TCS thrived on labor arbitrage. That model is dead. Generative AI writes code faster, cheaper, and with fewer bugs than a Level-1 engineer.

But-and this is a big but-from the ashes, a new beast is rising. A small, violent faction of startups is building world-class IP on shoestring budgets.

• SpaceTech: Look at Agnikul Cosmos and Skyroot. They are 3D-printing rocket engines and launching satellites for a fraction of SpaceX’s cost.

• AI: Look at Sarvam AI. They aren’t trying to out-English OpenAI; they are building “Sovereign AI” for India’s 20+ languages.

As Jensen Huang, CEO of Nvidia, told Indian leaders recently:

“India used to be a country of software export. In the future, India will be a country of AI export… You cannot just be the back office anymore. You must manufacture intelligence.”

The Oligopoly: The Feudal System of AI

Let’s be real. There is no “Free Market” in AI. There is a feudal system. You are a serf on digital land owned by five warlords. Economists call this the “Kill Zone” -where incumbents acquire or crush any threat before it can scale.

Nvidia is the Arms Dealer. Jensen Huang is the only man making guaranteed money in a gold rush because he’s selling shovels made of diamonds. Nvidia alone contributed to over 20% of the S&P 500’s returns recently. Jensen wasn’t subtle about it either: “We are producing something for the very first time… intelligence. We are at the beginning of a new industrial revolution.” Translation: “Pay me.”

Then you have the Warlords.

Microsoft is the Landlord. They own the enterprise distribution. If you build a B2B tool, Microsoft will just add it to Teams and kill you. Satya Nadella pulled off the coup of the century by outsourcing his R&D to Sam Altman, bypassing his own internal bureaucracy.

Google is the Awakened Giant. They were sleeping, woke up, panicked, and merged Brain and DeepMind into a singular war machine. If you have a “Search Startup” or an “AI Note-taking app,” you are squatting on Google’s property. They just released NotebookLM and Sherlocked half the productivity sector overnight.

Meta is the Joker. Mark Zuckerberg is the most dangerous man in AI because he doesn’t care about making money on the model. He’s releasing open-source Llama models for free just to burn the moats of OpenAI and Google. He wants to commoditize the intelligence layer so the only value left is attention (which he owns on Instagram).

And Apple? They are the Silent Assassin. They don’t talk about “AI.” They talk about “Apple Intelligence.” They will wait for you to invent a use case, watch you struggle, and then put it in the iPhone Control Center.

The math, however, is terrifying. OpenAI is valued at ~$300B. Anthropic at ~$18B+. David Cahn from Sequoia Capital calls it the “$600 Billion Question.” We need to see $600B of annual revenue to justify the capex. Right now? We are nowhere near that. We are burning cash to build a god, hoping the god pays us back before interest rates kill us.

The “Wrapper” Suicide Pact

Let’s talk about the “Thin Wrapper” startup. If your product is essentially OpenAI API + React Frontend, you are paying a “Stupidity Tax.”

Remember Jasper.ai? They were a unicorn printing money on copywriting. Then ChatGPT launched “Team” plans. Why pay a middleman when you can go to the source? It’s the “Feature, Not Product” trap. “Chat with PDF” went from a multimillion-dollar startup sector to a free button in Adobe Acrobat and Chrome in six months.

The AdTech world is facing an even weirder crisis: the “Dead Internet.” Advertisers are realizing they are paying for AI bots to click on ads written by AI bots. This is Baumol’s Cost Disease in reverse. Usually, services get more expensive over time. AI is causing the cost of cognition to plummet to zero, causing a deflationary spiral in the price of white-collar work.

As Satya Nadella put it: “The era of SaaS as we know it is coming to an end… AI becomes the central driver… The business logic is all going to these AI agents.”

The Financial Bloodbath

Wall Street has seen this movie before. They are now looking at Revenue Per Employee (RPE) with a microscope.

In the legacy Salesforce era, you were doing great at $200k revenue per employee. You hired armies of sales reps. In the AI-Native era of 2025, the benchmark is $750k — $1M+. Look at Midjourney-they generated ~$200M revenue with about 11 people. That is $18M per employee.

If you have 100 employees and only $5M in ARR, you are already dead. You just haven’t fallen over yet.

Predictions for 2026 are grim. We will see a massive wave of “acqui-hires” that are really just mercy killings. Big Tech won’t buy your company for the product. They will buy it for $10M just to get your 3 best AI engineers (PhD talent is still scarce), and then they will shut down your server the next day. The VC gets their principal back (maybe), the founders get golden handcuffs at Google, and the staff gets a “Thank You” email.

Vinod Khosla wasn’t joking when he said, “I literally tell people… I’m not going to fund somebody whose goal is to reduce risk. I’d rather live in a world where I don’t mind a high probability of failure.”

Future of Work: The Death of the “Feature Factory” & The End of Subscriptions

So, how do you survive? The “Feature Factory” model-where you hire 50 Product Managers to manage a backlog of minor tweaks-is over.

1. The “SaaS Depression” and the Subscription Revolt
In 2021, the average enterprise paid for 80+ SaaS subscriptions. “Digital Transformation” meant “Buying more logins.”
In 2026, CFOs are canceling everything. Nobody wants to pay $20/month per user for a tool that requires a human to operate it. We are moving from SaaS (Software as a Service) to Service-as-Software.

• Old Way: You pay $30/month for a tool, and you hire a human to use it.

• New Way: You pay the AI $100/month to do the job entirely.

• If your software doesn’t do the work, it’s just friction. The “workflow automation” tools like Zapier are being eaten by AI agents that can just navigate the web themselves.

2. The Fall of the Project Manager

The “PM-Industrial Complex” is collapsing. We used to have massive armies of Project Managers whose entire existence was to color-code Gantt charts, nag engineers for status updates, and “facilitate alignment.”

That era is over. AI generates the schedule. AI predicts the blockers. AI writes the status report before you even open the meeting.

Elon Musk brutally proved this at Twitter (now X) when he fired 80% of the staff-mostly middle management and coordinators-and the site didn’t crash; it shipped features faster. As Steve Jobs famously said years before the AI wave: “The doers are the major thinkers… the people that really create the things… are both the thinker and doer in one person.”

The “Middle Manager” who coordinates people is being replaced by the “Technical Founder” who coordinates APIs. The days of being a “non-technical” Project Manager are numbered. If you can’t read the code, deploy the build, or fix the bug yourself, you aren’t a manager anymore-you’re just friction. In 2026, if you aren’t building the product, you don’t have a job.

3. Team Size: The “Single Cookie” Rule
Jeff Bezos had the “Two Pizza Rule” (teams should be small enough to be fed by two pizzas).
The new rule is the “Single Cookie Rule.”

• Klarna replaced the workload of 700 customer support agents with one AI system.

• The “One-Person Unicorn”: We are approaching the era where a single developer, augmented by a fleet of AI agents (Devin, Cursor, Claude), can build a $1 billion company. The “10x Engineer” is now the “1,000x Engineer.”

• The Reality Check: You don’t need a Head of Engineering, a QA Lead, a DevOps Lead, and a Frontend Specialist. You need one obsession-driven polymath with an oversized monitor and a Cursor subscription.

4. Digital Transformation is Dead
Consulting firms like McKinsey and Deloitte used to sell “Digital Transformation” packages for $5M that took 3 years to implement.
Today, a founder can “transform” their business in a weekend by hooking up their database to an LLM. The “consulting arbitrage” is gone. Companies don’t want a 200-page slide deck on “Synergy”; they want an agent that auto-replies to 50,000 customer emails by Tuesday.
As Amjad Masad, CEO of Replit, predicted with chilling accuracy:

“The billion-dollar company with just one employee is coming. It’s not a matter of if, but when. We are compressing the time between ‘idea’ and ‘reality’ to near zero.”

The Clincher: My Moat is Fear

I saw Paul again a few months later.

The “Legal AI” startup was gone. The website now redirected to a 404 page. But Paul looked… good. Better than good. He was wearing a bespoke suit and an Apple Watch Ultra 2. We met at a private club, not a coffee shop.

“What happened?” I asked. “Did you raise a Series A?”

“I pivoted,” he said, deadpan.

“To what? Another wrapper?”

“God no,” he laughed, sipping a $25 aged scotch. “I’m a ‘Strategic AI Advisor’ now. I sit on the advisory boards of five different mid-sized legacy companies. Manufacturing, insurance, logistics. Boring stuff.”

“What do you actually do?” I asked.

He leaned in, a wicked glint in his eye. “I explain to 60-year-old board members and CXOs why they shouldn’t build their own LLMs. I sell them fear of missing out, and then I sell them ‘Consulting Services’ to do absolutely nothing.”

He held up five fingers. “Five companies. $10,000 a month retainer each. That’s $50k a month. My contract says I owe them ‘up to 10 hours’ of guidance per month. Do the math. I work 50 hours a month and make $600k a year.”

“And what is the advice?”

“I tell them: ‘Don’t build a model. You will lose. Just add a button on top of GPT-4. Focus on your data.’ I just repeat the same three slides about ‘The Kill Zone’ and ‘Data Gravity’ to terrified Boomers who think Skynet is coming for their pension.”

He sat back, satisfied. “It doesn’t matter if the advice works. It doesn’t matter if they succeed. They aren’t paying for results; they are paying for permission to ignore the hype, or permission to join it. I found my moat, my friend. My moat isn’t code. My moat is the insecurity of old rich men.”

The Lesson:
Technology changes. The hustle stays the same. Don’t build a wrapper. Build a bunker. Or better yet, be the guy selling the maps to the bunker. Because Sherlock is coming, and he doesn’t need a magnifying glass to see that your “moat” is just a puddle-unless you’re the one charging admission to see the puddle.

Originally published at https://www.amitgoel.me on December 11, 2025.

Sherlock Economics: Why Your $20M AI Startup Is Just an Unpaid Internship for OpenAI was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

Inspired by: Shreyas Doshi’s Deep Dive on Micromanagement. Listen here before you rage-quit: Understanding Micromanagement: A Deep Dive

There is a moment in Shreyas Doshi’s brilliant analysis where he drops a truth bomb that usually sends “visionary” Product Managers scrambling for their noise-canceling headphones. He essentially argues that micromanagement is not a binary evil; it is a situational necessity.

To paraphrase one of his core insights: “High-performance execution often looks like micromanagement to low-performance employees.”

We treat micromanagement like it’s a contagious skin disease. But I’m here to tell you-as a seasoned Product Manager, entrepreneur, and engineer with 25+ years of scars-that most of the times you * do NOT need* a boss who breathes down your neck. But sometimes you do need one.

I have written code for Pocket PCs. I have built encryption systems for software and middleware for satellite PayTV systems (you know, those clunky boxes that enabled you to watch linear TV a decade ago-do they even still exist? Or are they just in a landfill somewhere wondering where the signal went?). I have ported WebKit browser engines into set-top boxes with 64MB of RAM and built Server-Side Ad Insertion (SSAI) for streaming.

I have seen the industry change from “How do we save a kilobyte?” to “Let’s just spin up another AWS instance.” And through it all, one truth remains: A fine balance is a myth. Execution is everything.

Sometimes, micromanagement is the only thing standing between a successful launch and a spectacular fireball. And sometimes, you have a boss who is just a control freak disguised as a “coach.” The trick is knowing the difference.

The Good Hover: The Day the Palm Pilot Died

Let’s rewind to 2001. “Cloud” meant rain was coming, “Agile” was a gymnastics term, and I was writing C code for the Palm Pilot.

For the uninitiated, the Palm Pilot ran on the DragonBall processor and had about as much memory as a modern refrigerator’s ice maker. If you messed up a pointer, you didn’t just get an error message; you got the “Fatal Exception” screen of death, and you had to find a paperclip to reset the device physically.

I was building a sync conduit-a piece of middleware supposed to talk to a desktop PC. It was crashing. Constantly.

My manager, let’s call him The Colonel, rolled his chair over.

Now, according to modern HR theory, this is an invasion of my “psychological safety.” According to Self-Determination Theory (SDT), humans crave autonomy, and The Colonel was threatening that. He didn’t ask me how I felt about the crash. He didn’t offer me a mental health day.

He said, “Open the memory handle lock. Show me the structure packing. Now.”

For four hours, he sat there. He watched me type. He pointed at the screen with a finger that smelled of cheap coffee. “You’re locking the memory chunk but not unlocking it before the serial port interrupt fires. You’re fragmenting the heap.”

I hated him. I felt suffocated. My Amygdala was firing off “fight or flight” signals because my ego was being crushed.

But here is the behavioral science twist: This was Task-Relevant Maturity (TRM) in action. As Andy Grove famously put it, when TRM is low (I was a junior idiot), the management style must be structured and hands-on.

We fixed the bug. The sync worked. The product shipped.

If The Colonel had “empowered” me to figure it out myself, we would have missed the launch, and I would have been fired. He didn’t micromanage me because he didn’t trust me as a person; he micromanaged the process because the cost of failure was absolute.

“My job is not to be easy on people. My job is to make them better.” — Steve Jobs

The Bad Hover: The “Helicopter Founder” (A Horror Story)

Now, let’s flip the coin. Let’s talk about The Founder Who Can’t Let Go.

This is the dark side. I once worked for a Founder-let’s call him Steve-Not-Jobs.

Steve hired a team of senior engineers, ex-Google architects, and veteran PMs. We had 100 years of combined experience. We knew how to build scalable SaaS.

But Steve had what psychologists call the Illusion of Control. He believed that if he wasn’t in the room, the decision wouldn’t be made “correctly.”

• The Symptom: Steve would join every meeting. Even the stand-ups. Even the code reviews.

• The Excuse: “I just want to be hands-on! I’m a product-first CEO! I’m coaching you guys!”

• The Reality: He was treating highly capable experts like interns.

This triggers Learned Helplessness. When a team realizes their decisions will always be overturned by Steve, they stop making decisions. They stop innovating. They just wait for Steve to tell them what color the button should be.

I remember proposing a detailed architecture on how the system will work. Steve interrupted the meeting to argue about the font size on the login page for 45 minutes. He called it “attention to detail.” We called it “why are we here?”

This isn’t situational leadership. This is Narcissistic Supply. The founder uses the team as an audience to perform his own brilliance. He doesn’t want the product to succeed; he wants to be the reason the product succeeds.

“If you want to hire great people and have them stay working for you, you have to let them make a lot of decisions and you have to be run by ideas, not hierarchy.” — Steve Jobs (ironically, the real one).

The HR Blind Spot: Why They Always (Okay! sometimes) Get It Wrong

Here is where the tragedy turns into comedy. Human Resources (HR) is statistically incapable of distinguishing between The Colonel (Good Micromanagement) and Steve-Not-Jobs (Bad Micromanagement).

Case A: The False Positive (Punishing the Good Manager)

I have seen HRBPs destroy great engineering cultures.

Let’s say I have a PM named Kevin (we all know a Kevin). Kevin loves strategy but hates execution. Kevin refuses to write detailed JIRA tickets because “that’s tactical.” When I start “micromanaging” Kevin-checking his work, demanding specs, sitting in his meetings to ensure he isn’t lying to customers-Kevin runs to HR.

Kevin: “I feel unheard. My manager is toxic.”
HR: “Oh no! We must protect psychological safety!”

HR applies the Fundamental Attribution Error. They assume I am a bad person (controlling) rather than looking at the situation (Kevin is incompetent). They put me on a Performance Improvement Plan for “leadership style,” while Kevin continues to produce slide decks that mean nothing.

Case B: The False Negative (Enabling the Bad Founder)

Conversely, HR often protects the Steve-Not-Jobs types.

When employees complain that Steve is suffocating them, HR frames it as: “He’s just so passionate! He’s a visionary! You need to learn to manage up!”

They mistake Neuroticism for Passion. They view his intrusion as “high engagement.”

This gaslights the employees into thinking they are the problem, leading to massive burnout and turnover.

“Success breeds complacency. Complacency breeds failure. Only the paranoid survive.” — Andy Grove

The “Strategy-Only” PM: Every Team Has One

Let’s go deeper on Kevin.

Every organization has a “Strategy-Only” PM. This person believes they are the “CEO of the Product.” They want to set the vision, do the podcast circuit, and write “Thought Leadership” posts on LinkedIn.

But ask them to test the product on a staging server?

Kevin: “I’m really more about the why, not the how. I don’t want to get into the weeds.”

I have spent years fighting Kevins. The behavioral science here is the Dunning-Kruger Effect-a concept I explored in depth in my previous article on the Dunning-Kruger Effect: The Blind Spot of a Product Manager.

Kevin is so incompetent at execution that he literally cannot recognize how bad he is at it. He thinks “Strategy” is a separate output from “Execution.”

But as I learned building those satellite TV boxes: Strategy is just a hallucination until the encryption key validates.

When I force Kevin to do the grunt work, he experiences Cognitive Dissonance. He sees himself as a visionary; I treat him like a worker bee. To protect his ego, he labels me a “Micromanager.”

If you have a Kevin on your team, you must micromanage him. You have no choice. Because if you don’t, he will sell a vision to the stakeholders that your engineering team cannot build, and you will be the one left holding the bag when the deadline explodes.

“The smartest people are constantly revising their understanding, reconsidering a problem they thought they’d already solved. They’re open to new points of view, new information, new ideas, contradictions, and challenges to their own way of thinking.” — Jeff Bezos

The AI Apocalypse: Judgment Day for the “Ideas Guy”

And now, the final boss: Artificial Intelligence.

If you thought micromanagement was a hot topic before, wait until AI really lands. The discourse is about to shift from “Who does the work?” to “Who verifies the truth?”

1. The Value Shift: From Creation to Judgment

We are entering an era where “creating” is free. You want code? Claude gives it to you. You want a PRD? ChatGPT writes it. You want a roadmap? Gemini sketches it.

This kills the “Idea Guy.” The only value left is Judgment and Execution.

• The Trap: AI is a confident liar. It will write a SQL query that looks perfect but deletes your production table. It will write a legal clause that sounds professional but violates GDPR.

• The New Skill: Can you smell the lie? Can you look at the code generated by the AI and know that it’s introducing a security vulnerability? Can you take the strategy the AI proposed and execute it against the messy reality of legacy systems?
  If you cannot Judge the output and Execute the implementation, you are just a middleman between a hallucinating robot and a frustrated customer.

2. The Hallucinating Founder (Fintech Edition)

We are already seeing founders who use ChatGPT as their CTO.

• Founder: “I asked ChatGPT, and it said we can integrate with these three major banks and process a million micro-payments per second by next Friday.”

• Me: “The AI assumes banks use modern REST APIs. In reality, that bank runs on a mainframe from 1982 using COBOL and batch files that only update at midnight. The AI is hallucinating a world that doesn’t exist.”
  This is the new danger. The founder micromanages the team based on a fantasy timeline created by an LLM that has never had to wait for a bank’s compliance department to approve a firewall rule.

3. AI as the “Positive Micromanager” (The New Hope)

However, there is a silver lining. AI enables “Async-Micromanagement” -the holy grail where rigor is enforced without the emotional toxicity of human hovering.

Here is how AI will positively transform the landscape:

• The Depersonalized “Bad Cop”
  Normally, when I tell Kevin his PRD is vague, he takes it personally. But if I configure an AI Agent with strict rubrics (“Must contain error states,” “Must define API latency limits”), and Kevin submits his work to the AI first, the AI does the nitpicking.

The Result: The AI flags the missing edge cases. Kevin fixes them to get the “green light.” By the time the doc reaches me, it’s 90% solid. The AI did the micromanaging, and I get to be the strategist.

• The Socratic Simulator
  Before a high-stakes review, I can tell my team: “Simulate the meeting with this specific AI Persona.”
  They can upload their strategy to a GPT prompted to act like a cynical CTO. The AI will tear their logic apart, finding holes in their reasoning.

The Result: The team gets “micromanaged” in a safe sandbox. They fix their own gaps. When they present to me, they are bulletproof. This builds Task-Relevant Maturity faster than I ever could manually.

• Observability vs. Surveillance
  Micromanagement often stems from a lack of visibility. “What is Kevin doing?”
  AI tools can now ingest Git commits, Jira updates, and Slack threads to generate a “Pulse Report”-not to spy, but to summarize progress.

The Result: Instead of tapping Kevin on the shoulder every 3 hours asking “Is it done?”, I look at the AI summary. I see the blockers. I see the velocity. I stay informed (micromanagement needs satisfied) without interrupting the workflow (autonomy preserved).

• The “Prompt-Based” Standard
  Micromanagement is often just an attempt to enforce a standard of quality. With AI, you can encode that standard into a Prompt Library.

The Result: “Don’t just write a user story; use this prompt that forces you to consider the mobile constraints we discussed.” The prompt is the micromanager. It guides the execution path step-by-step, ensuring the output aligns with the vision, without me having to stand there reciting the rules.

“Artificial Intelligence is no match for Natural Stupidity. But together? They are unstoppable.” — Tech Industry Proverb

Conclusion: Autonomy is Earned, Not Given

So, let’s wrap this up with a closure to our narrative.

Years after the Palm Pilot project, I ran into The Colonel at a conference. I bought him a beer.

“You were a nightmare,” I told him. “You hovered over me like a drone strike.”
He laughed. “And did the code crash?”
“No.”
“Did you learn how memory heaps work?”
“Yes.”
“Then I wasn’t micromanaging you,” he said. “I was teaching you how to survive.”
This is the message.

If you are a leader: Do not apologize for the hover. If the stakes are high and the person is unproven, get in the weeds. If they are a “Kevin,” manage them out or manage them up, but do not let them hide behind “strategy.”

If you are an employee: Stop whining about micromanagement and start shipping. The fastest way to get your boss off your back is to execute so flawlessly that they get bored watching you.

Autonomy is not a right found in the constitution. It is a reward for competence.

Now, if you’ll excuse me, I need to go micromanage my toaster. I don’t trust its “browning settings” one bit.

One Final Thought:”It is a grave error to think that delegation is the opposite of micromanagement. There is no way you can micromanage everything… but you must selectively micromanage the things that matter.” This video is relevant as it features Shreyas discussing the different altitudes of leadership and how misapplying them leads to the “under-management” trap described above.

For more context on the leadership styles discussed here, specifically the nuances of “The Captain” vs “The General,” check out this discussion: Coaching Product Leadership with Shreyas Doshi: Part 2

Originally published at https://www.amitgoel.me on December 9, 2025.

The Art of the Hover: The Good, The Bad, and The Automated-Why AI Is the ‘Positive Micromanager’… was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

“The first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency.” — Bill Gates

This is Part 3 of my series on the collapse of the traditional software industry.

• Part 1: How “Vibecoding” is Killing the SaaS Dinosaurs (Production is free).

• Part 2: Stop Charging Rent to Robots (Pricing is broken).

• Part 3: Today, let’s talk about how AI is going to disrupt your organization design and strategy

I recently had coffee with two founders in Singapore, and the contrast explained everything you need to know about the next decade of tech.

Founder A was beaming. He just closed his Series A. “We’re scaling up,” he told me, puffing out his chest. “We’re moving into a new office in the CBD. I just hired a VP of People, a VP of Strategy, and we’re opening a new engineering hub in Bangalore. We’re going to hit 80 employees by Q4.”

Founder B looked confused. She runs a company with similar revenue ($3M ARR), but she looked rested. “We just hired our third employee,” she said. “He’s a full-stack architect who manages our swarm of AI agents. I think we might need one more person for legal compliance, but otherwise, we’re good.”

Founder A is building an Empire. Founder B is building a Fortress.

In the old world, Founder A wins. In the AI world, Founder A is dead. He just built a massive, high-friction organization that will be outmaneuvered by Founder B’s three people and 50,000 GPUs.

There is a comforting lie floating around LinkedIn right now. It goes: “AI won’t replace you; a person using AI will replace you.”

It’s a nice sentiment. It suggests that everyone gets to keep their job, just with better tools. I2t implies that the junior developer who struggles to center a div will suddenly become a 10x engineer just because they have Copilot.

This is mathematically impossible. If you multiply Zero by 100, you still get Zero.

AI is not a magic wand that bestows competence on the incompetent. It is a Force Multiplier. It amplifies the intelligence of the user. If you are a brilliant architect, AI makes you a god. If you are a mediocre “ticket taker” who just patches bugs, AI makes you obsolete.

We are witnessing the end of the “Code Monkey” era, the collapse of the Global Labor Arbitrage model, and the rise of a new corporate metric: Profit-Per-Employee.

1. From “Specialist” to “God-Tier Generalist”

For the last decade, we fetishized hyper-specialization. We broke engineers into tiny fragments: “React Frontend Developers,” “Postgres DB Admins,” “DevOps Engineers.”

This fragmentation created a massive Inefficiency Tax. A simple feature required a meeting between the Frontend guy, the Backend guy, and the DB guy. It required a Project Manager to coordinate them. It required a Jira board to track them.

But in the Vibecoding era, the Specialist isn’t just dying-they are evolving.

The AI removes the barrier of Syntax. The Backend Engineer no longer needs to memorize CSS flexbox rules; the AI handles that. This allows a high-competence engineer to escape their silo. They are no longer just a violinist; they are the conductor and the entire orchestra.

“Specialization is for insects.” — Robert Heinlein

If you are a developer who defines your value by “knowing React” rather than “solving business problems,” you are in trouble. The syntax is free. The value has shifted entirely to System Architecture, Data Modeling, and Taste.

2. The Death of Human Middleware (Baumol’s Cost Disease)

The casualties won’t just be engineers. The “Non-Technical Middle Manager” is walking dead.

This is an economic phenomenon known as Baumol’s Cost Disease. Usually, this theory explains why services (like education) get more expensive while goods (like TVs) get cheaper. In Tech, the “Good” (Code) is dropping to zero cost. The “Service” (Management) remains expensive. The market will correct this imbalance by eliminating the management layer entirely.

The Non-Technical Product Manager

We all know the type. The PM who is a “Domain Expert” but doesn’t understand how the sausage is made. They write vague user stories like “As a user, I want the dashboard to load faster,” and then they wait for Engineering to tell them if it’s possible.
In an AI world, this PM is useless. Why? Because you cannot prompt an AI if you don’t understand the architecture. If you ask an AI agent to “build a dashboard,” it will build garbage. You need to tell it: “Build a React dashboard using Recharts, pulling from the Snowflake API, caching via Redis to reduce latency.”

The End of the SDR and Onboarding Manager

• Sales Development Reps (SDRs): The job of sending 100 generic emails a day is dead. An AI agent can send 10,000 hyper-personalized emails, read the responses, and book the meeting. The “Volume Game” belongs to bots now.

• Customer Success: Why are we paying humans to teach other humans how to use software? AI agents will guide users through the product in real-time. The “Onboarding Manager” reading from a script is redundant.

3. The Economics of Labor Arbitrage (Ronald Coase & The Transaction Cost)

This brings us to the most uncomfortable reality: The collapse of the Outsourcing Model.
For 20 years, the software industry relied on Labor Arbitrage. The math was simple:

• The US Reality: An engineer in San Francisco costs $200,000.

• The Offshore Reality: An engineer in Bangalore costs $30,000.

• The Arbitrage: You hire the offshore team. Even if they are slower, the price difference is so massive you tolerate the inefficiency.

AI breaks the arbitrage math.

We can explain this using Ronald Coase’s Theory of the Firm. Coase argued that firms exist to reduce Transaction Costs (search, coordination, contracting). Outsourcing increases transaction costs (communication latency, cultural context, quality control), but low wages offset it.

AI is a massive deflationary force that eliminates the transaction costs.

• Old Math: 1 Hired Engineer ($200k) vs. 5 Offshore Engineers ($200k).

• New Math: 1 Hired Engineer + AI ($300k) = 10x Output.

Suddenly, the single US engineer is more productive than the team of 5, but without the Transaction Costs-no timezone delays, no language barriers, no “Project Manager” layer.

“The most expensive thing in software is communication. If you can eliminate the need to communicate by having one person do the work of ten, you win.” — Naval Ravikant

The “Managed Services” model (Infosys, Wipro) is the next Blockbuster. They sell “seats” and “man-hours” for maintenance and testing. But AI does maintenance and testing instantly for free. If your business model depends on billing hours for work that requires little creativity, you are technically a “human API,” and you are about to be deprecated.

4. The New Scoreboard: Profit-Per-Employee (PPE)

Because of this, the metric for success is flipping. For decades, founders like Founder A measured their worth by the size of their “Empire” (Headcount). “We just crossed 100 employees!” was a badge of honor.

In the AI era, Headcount is a liability. The smart money is watching Revenue-Per-Employee (RPE).

• Legacy SaaS Co (Founder A): $3M ARR / 80 Employees = **$37,500 RPE**.

• AI-Native Co (Founder B): $3M ARR / 3 Employees = **$1,000,000 RPE**.

The AI-Native company can undercut the Legacy company on price by 50% and still be more profitable. They can spend 10x more on customer acquisition.

The “Throwing Bodies at the Problem” Fallacy (Brooks’s Law)

Legacy companies are addicted to Linear Scaling. When a problem arises (e.g., “Support tickets are piling up”), their reflex is to treat it like a manufacturing problem: Throw more bodies at it.

Step 1: Hire 20 more support agents.
Step 2: Realize 20 people create chaos.
Step 3: Hire 2 Managers to manage them.
Step 4: Hire a Director to manage the Managers.
You just built a pyramid of human routers.

This triggers Brooks’s Law: “Adding manpower to a late software project makes it later.” In business terms, this is Diseconomies of Scale. Each new human adds “Communication Overhead.” As you add people, the complexity of the network increases exponentially (N(N−1)/2), slowing down decision-making.

This is Organizational Debt. Just like technical debt, you are borrowing against future speed to solve a problem today with the wrong tool (humans).

In the AI era, this is suicidal. When that same problem arises in an AI-native company, they don’t open a job requisition. The Full-Stack Engineer fine-tunes an LLM on the last 10,000 tickets.

• Cost: $50 in compute.

• Time: 2 hours.

• Headcount Added: Zero.

• Result: The AI-native company scales exponentially (Software economics), while the Legacy company scales linearly (Service economics). The Legacy company will eventually be crushed by its own weight.

5. The Pivot: If Building is Free, What Do We Sell? (Liability-as-a-Service)

This leads us to the existential crisis facing the software industry. We are rapidly approaching a zero-marginal-cost reality for code generation. If a small team-or even a single “non-technical” founder-can prompt an AI agent to “vibecode” a functional Salesforce clone in a weekend, the fundamental value proposition of SaaS is shattered.

Why would an enterprise customer pay Salesforce $300 per seat/month for software that a college student just replicated on their laptop for the cost of an API token?

The answer lies in a shift from Capability to Accountability. The answer is Trust.
We are leaving the era of Software-as-a-Service (SaaS) and entering the era of Liability-as-a-Service (LaaS).

The Code is a Commodity; The Shield is the Product

In this new paradigm, the ability to write code is table stakes. It is abundant, cheap, and accessible. You can direct an AI to build a payroll application in 48 hours. It will have a sleek UI, it will connect to bank APIs, and it will run perfectly… until it doesn’t.

The “product” is no longer the software functioning correctly when things go right; the product is what happens when things go wrong.

• The Payroll Scenario: Imagine you use a custom, AI-generated payroll script. It works for six months. Then, the AI “hallucinates” during a routine update-perhaps misinterpreting a new state tax code-and fails to withhold taxes for 500 employees. The Result? You, the employer, are liable. The IRS does not care that your AI hallucinated. You face audit penalties, lawsuits from employees for back-taxes, and potentially jail time for gross negligence.

• The Workday Scenario: Conversely, when an enterprise pays Workday or ADP millions of dollars, they are not paying for the SQL database that stores employee names. They are buying an insurance policy. If Workday glitches and messes up tax withholdings, Workday pays the fine. Workday deploys an army of lawyers and accountants to fix it.

The “Throat to Choke”

For the Fortune 500, software purchasing decisions are driven by risk mitigation, not just feature acquisition. This brings us to the concept of the “Throat to Choke.”

In corporate governance, there is a massive premium placed on shifting blame. If a CIO buys a custom AI solution from a nimble startup and it causes a data breach, the CIO is fired for recklessness. If the CIO buys Microsoft or Oracle and the same breach happens, it is considered an “unfortunate vendor incident.” The CIO keeps their job because they made the “safe” choice.

Real-World Example: The CrowdStrike Outage

Consider the massive CrowdStrike outage of 2024. When a faulty update crashed millions of computers globally, it cost Delta Airlines alone over $500 million.

• Scenario A (The Startup): If that update had come from a cheap, AI-generated endpoint protection tool, Delta would have had no recourse. The startup would have simply declared bankruptcy, leaving Delta with the bill.

• Scenario B (The Incumbent): Because it was CrowdStrike, a massive public entity, Delta had a target for litigation. The “product” CrowdStrike sells is not just antivirus; it is the capital reserves and insurance policies required to absorb the shock of failure.

The Moat of the Legacy Dinosaur

The “Moat” for legacy giants-the Salesforces, SAPs, and Epics of the world-is no longer their source code. Their code is often legacy spaghetti that is worse than what an AI could write today.

“Trust is the coin of the realm.” — George Shultz

Their moat is their Indemnity Shield.

1. SOC2 & ISO Compliance: These are grueling, expensive, human-centric auditing processes that prove a company handles data safely. An AI agent cannot “vibecode” a SOC2 Type II audit report.

2. Regulatory Navigation: In healthcare (HIPAA) or finance (PCI-DSS/GDPR), the software must adhere to laws that change frequently. Legacy giants have armies of humans monitoring regulatory changes to update the compliance layer of the software.

3. Financial Backstop: Enterprises buy software to offload risk. They pay a premium for the legal guarantee that if something breaks, it’s not the CTO’s fault.

Conclusion: Don’t Build an Empire, Build a Fortress

To the founders reading this: Stop trying to hire your way to success. Stop measuring your self-worth by the size of your All-Hands meeting.

In the previous era, we built Empires. We hired armies of humans because human labor was the only way to scale output. An Empire is impressive to look at, but it is cognitively fragmented. Intelligence is fragmented across hundreds of nodes (people), slowing down the collective brainpower to the speed of a calendar invite.

In the AI era, you must build a Fortress.

A Fortress isn’t just about automation; it’s about concentrated intelligence. It is not defined by how many hands are typing, but by the clarity of the mind directing them.

The difference is physics.

• The Empire operates on Distributed Intelligence. It relies on the consensus of the many, moving at the speed of biology (meetings, emails, persuasion).

• The Fortress operates on Amplified Intelligence. It relies on the judgment of the few, amplified by silicon, moving at the speed of thought.

The Tale of Two Futures

Let’s go back to those two founders one last time.

Founder A (The Empire) is currently in a 4-hour “strategy offsite” in Singapore. He is mediating a dispute between his VP of Product and his VP of Engineering about “roadmap alignment,” while his offshore team in Bangalore waits 12 hours for instructions. He is burning $400k a month to generate $3M in value. He feels important, but he is actually just a high-paid babysitter for a bloated organization.

Founder B (The Fortress) is at the beach. She isn’t checking Slack because there is no Slack. Her agents are executing the strategy she designed last week. She isn’t working in the machine; she is designing the machine. She is burning $40k a month to generate the same $3M in value.

The Reality Check

The market is about to run a ruthlessly efficient garbage collection algorithm on the tech sector.

For the last ten years, we confused “Headcount” with “Brainpower.” We thought that if we hired 100 mediocre engineers, we would get the output of 10 geniuses. We were wrong. We just got 100 times the complexity.

The next Amazon or Google will not have 100,000 employees. It might not even have 1,000. It will be a small, terrifyingly efficient Fortress of “God-Tier” Generalists using AI not just to write code, but to simulate futures, predict risks, and execute decisions instantly.

In the future, there are only two roles left:

1. The Architect who provides the judgment.

2. The Artifact that executes the work.

The asteroid is already here. Stop building for the dinosaurs.

Originally published at https://www.amitgoel.me on December 7, 2025.

AI Is Only As Intelligent As The Person Using It : Why Common Sense and Expertise Are the New Moat was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

This is Part 2 of my series on the collapse of the traditional software industry. In Part 1, How “Vibecoding” and The Outcome Economy Are Killing the SaaS Dinosaurs, I explained how AI is making software production free. Today, let’s talk about why your pricing model is about to bankrupt you.

If you read the first post, you know we are witnessing the extinction event of the “SaaS Dinosaur”-those bloated software companies that charge you rent to move data from Column A to Column B.

But there is a second asteroid hitting the industry, and this one isn’t about code. It’s about money. Specifically, the collective hallucination known as Seat-Based Pricing.

“Software is eating the world.” — Marc Andreessen (2011)

Marc was right. But in 2025, he forgot to mention that the software would eventually eat the business model that paid for it.

I recently sat in on a Zoom call that perfectly encapsulates the absolute delusion of the current software industry.

On one side was an “Enterprise Account Director” from a massive, publicly traded AdTech company-let’s call the company OptiMax. The rep, let’s call him Jason, had hair so perfect it looked rendered in Unreal Engine 5. He was wearing a Patagonia fleece vest that had clearly never encountered nature.

On the other side was the CTO of a fast-growing D2C e-commerce brand. She looked tired. She looked like someone who actually ships code.

Jason was trying to close a renewal for their Creative Management Platform. He flashed a blindingly white smile and said, “So, to support your hyper-growth next year, we recommend bumping you up to the ‘Enterprise Synergy’ tier. That unlocks 40 seats for your design team, unlimited folder nesting, and our new ‘Collaboration Lounge’.”

The CTO rubbed her temples. You could physically see her soul leaving her body to go to a happier place.

“Jason,” she said. “We don’t have 40 designers anymore. We have two. And we have an AI agent swarm running on a local server that generates 5,000 banner variations a week, tests them on Meta, and kills the losers automatically.”

Jason blinked. His brain, trained on a decade of “Land and Expand” playbooks, short-circuited. “Okay… well, that sounds… disruptive. But the Enterprise tier requires a minimum of 20 seats to unlock the API rate limits. Can you… can you assign seats to the agents? Like, create email addresses for the Python scripts?”

The CTO stared at him. The silence was so loud you could hear Jason’s career evaporating.

“You want me,” she said slowly, “to create fake email addresses for a script named banner_bot_v4, just so you can charge me $150 a month for it to not log in, not look at your beautiful dashboard, and not attend your webinar on ‘The Future of Creativity’?”

“Well,” Jason stammered, mentally calculating the hit to his Q4 commission. “How else are we supposed to bill you? It’s industry standard.”

The CTO didn’t renew. She hired a freelancer to build a lightweight interface on top of the Midjourney API and a headless CMS. Jason lost the deal. OptiMax lost the client.

This is the state of B2B SaaS in 2025. We are trying to sell gym memberships to robots that don’t have bodies.

The Seat-Based Ponzi Scheme (Zero Marginal Cost)

For the last fifteen years, the “Per User/Per Month” subscription model has been the golden calf of the tech industry. Venture Capitalists loved it. It was predictable. It relied on the economic concept of Zero Marginal Cost. Once the software is built, it costs the vendor $0.00 to add another user, so that $50/month fee is pure profit margin.

“Show me the incentive and I will show you the outcome.” — Charlie Munger

Munger’s wisdom explains why we are here. The model worked because humans were the bottleneck. If you wanted to scale your ad campaigns, you had to hire more traffickers, designers, and copywriters. If you hired more people, you bought more seats. Revenue grew linearly with headcount.

But we have entered the age of AI. The goal of every competent company today is to reduce headcount while increasing output.

If your pricing model relies on my headcount growing, you are taking a short position on technology. You are betting that I will fail to automate.

If I successfully replace my 20-person ad ops team with one really smart LLM agent, your revenue drops from $3,000/month to $150/month. This creates a perverse incentive structure: You are financially incentivized to keep my business inefficient.

That is not a partnership. That is a hostage situation.

The Great Rate Card Scam (Price Discrimination)

Before we talk about the future, let’s look at the “Menu of Confusion” that most SaaS companies are still using today. Economists call this Price Discrimination: the art of charging different customers different prices for the exact same product, based on their willingness to pay rather than the cost to serve.

It usually looks like this:

1. The “Starter” Tier ($0 — $20): This exists only to mock you. It allows you to create a project, but if you want to export it, save it, or actually use it, a popup appears telling you to upgrade. It is software as a ransom note.

2. The “Pro” Tier ($50 — $100) — ⭐️ RECOMMENDED: This is the trap. They highlight it in blue. They put a “Most Popular” badge on it. It includes 5 seats you don’t need, but it holds the one feature you do need (like API access) hostage.

3. The “Enterprise” Tier (Contact Sales): “Contact Sales” is Latin for “How much money do you have?” This tier is identical to the Pro tier, except it includes Single Sign-On (SSO).

“Enterprise software pricing is about extracting the maximum amount of money from the customer without them laughing in your face.” — Jason Fried (CEO, Basecamp)

This model relies on Feature Gating. The value isn’t in using the product; the value is in unlocking the gate.

But in an AI world, this collapses. If I am using an AI Agent to interact with your software via API, I don’t care about your UI features. I don’t care about your “Dark Mode.” I don’t care about your “Collaboration Lounge.” I just want the data in and out.

The “Long Rate Card” becomes irrelevant when the user is a Python script. A Python script doesn’t have an ego, and it doesn’t need a personalized dashboard.

The Pivot: Usage-Based Pricing (The Physics of Value)

The companies that will survive are the ones shifting to Usage-Based Pricing (UBP).

This isn’t just “Pay As You Go.” It is a sophisticated metering of value consumption. It shifts the metric from “Access” (Login) to “Throughput” (Work Done).

“Your margin is my opportunity.” — Jeff Bezos (Founder, Amazon)

Bezos understood this before anyone. AWS didn’t charge you for “Server Access.” They charged you for the seconds your server ran. This aligned incentives perfectly: if you grew, Amazon grew.

How It Actually Works: Monitoring & Measurement

In a UBP model, the software vendor installs a digital “utility meter” inside the codebase.

• Metering Events: Every time the software performs a distinct action (e.g., an image is resized, a query is run, a fraud check is passed), a “metering event” is fired to the billing engine.

• Aggregation: These events are aggregated. You don’t see 10,000 line items; you see “10,000 units consumed.”

Real World Examples

1. The AI Infrastructure Model (OpenAI / Anthropic)

• The Metric: Tokens.

• Why: They don’t charge you for “having an API key.” They charge for the complexity of the thought. An essay costs more than a tweet. This is pure physics-based pricing. If you stop thinking, you stop paying.

2. The New Age AdTech Model (The “Vibecoded” Ad Server)

• Old Way: Monthly fee based on “Estimated Impressions” tiers. (Guesswork).

• New Way (UBP): Rendered Creatives. The system counts exactly how many times the AI generated a unique banner ad variation. If the client’s campaign is paused, their bill drops to near zero. If they scale for Black Friday, the bill spikes, but they don’t mind because they are printing money.

The vendor becomes a partner in efficiency, not a landlord of shelfware.

The Innovator’s Dilemma: Why Jason Is Mathematically Prohibited From Changing

Here is where the sarcasm needs to take a backseat to cold, hard economic theory. The legacy SaaS companies-the Salesforces, the Adobes, the OptiMaxes-are in serious trouble. And it’s not because they are stupid. It is because they are trapped in what Clayton Christensen famously called The Innovator’s Dilemma.

“The reason why it is so difficult for existing firms to capitalize on disruptive innovations is that their processes and their business model that make them good at the existing business actually make them bad at competing for the disruption.” — Clayton Christensen (The Innovator’s Dilemma)

For Jason at OptiMax, the problem isn’t just outdated pricing; it is that his Pricing Model dictates his Product Strategy. This is a financial version of Conway’s Law: systems are designed to mirror the communication structures of the organization.

In SaaS, your revenue model designs your roadmap. Let’s break down the two opposing physics at play here.

1. The Seat-Based Trap: Optimizing for Friction (SaaS 1.0)

When you charge per seat, your North Star metric is Retention. To keep a human paying $100/month, that human must feel “locked in.”

• The Incentive: The Product Manager is incentivized to build “Stickiness.”

• The Feature Set: Complex dashboards, social feeds, gamification, and proprietary file formats.

• The Hidden Goal: Maximize Time-on-Site. If the user finishes their work in 3 seconds and leaves, they might question why they pay for the subscription. So, the software subtly adds friction to justify its existence.

2. The Usage-Based Future: Optimizing for Flow (SaaS 2.0)

When you charge for usage (compute/throughput), your North Star metric is Consumption. To keep a machine paying $0.05/transaction, that machine must encounter zero resistance.

• The Incentive: The Product Manager is incentivized to build “Velocity.”

• The Feature Set: High-performance APIs, Webhooks, SDKs, and Headless architectures.

• The Hidden Goal: Minimize Time-to-Outcome. If the task takes 3 seconds, you optimize it to 0.3 seconds so the client can run it 10,000 times more.

The “Cannibalization” Death Spiral

This brings us to the specific mechanism that kills the dinosaur. I call it the Revenue Cannibalization Loop. Legacy companies cannot switch to usage pricing because the math doesn’t work in reverse. They are addicted to the “inefficiency margin.”

Imagine you are the CEO of OptiMax. You charge $100/seat.
Your engineering team proposes a new “AI Auto-Resize” feature that automatically formats banners for Instagram, TikTok, and LinkedIn.

• The Efficiency: It saves the client 100 hours a week.

• The Economics: If you ship this feature, your client fires the 5 junior designers whose only job was resizing banners.

• The Consequence: If they fire those designers, they cancel 5 seats ($500/month revenue loss).

This is the definition of a Perverse Incentive.

• If you innovate (automate work): You lose revenue immediately.

• If you stagnate (keep manual work): You lose the customer eventually to a disruptor.

Jason chooses stagnation every time. He has a quarterly quota to hit. He cannot sell a feature that shrinks his deal size.

The Future: The “Outcome Economy” (Next 5 Years)

So, where does this go?

In the short term (1–3 years), we will see a messy transition to Hybrid Pricing. Companies will charge a small platform fee plus a “consumption tax” on AI tokens. It’s clunky, but it stops the bleeding.

But the real endgame (5 years out) is Outcome-Based Pricing. This is where “Agency-as-a-Service” becomes real. This solves the Principal-Agent Problem. In the old model, the software (Agent) didn’t care if the user (Principal) succeeded, as long as they paid the bill. In the new model, incentives are aligned.

“We are moving from a world of ‘software tools’ to a world of ‘digital employees’. You don’t buy a tool; you hire an agent.” — Vinod Khosla (Khosla Ventures)

If I hire an AI Media Buyer, I shouldn’t pay for the software or the usage. I should pay for the Result.

• Seat Model (Old): Pay a DSP (Demand Side Platform) a monthly fee + % of spend to access the dashboard. (You take the risk).

• Usage Model (Current): Pay $0.01 per thousand bid requests processed. (Shared risk).

• Outcome Model (Future): Pay a % of the ROAS Lift or a fixed fee per Converted Customer. (Vendor takes the risk).

In an outcome model, the software vendor is incentivized to be ruthlessly efficient. If their internal AI agent wastes 1,000 tokens trying to find a customer, they eat that cost, not me. This drives massive innovation in efficiency.

The vendor becomes a partner. They aren’t selling you a shovel; they are selling you a hole. And if the hole isn’t dug, they don’t get paid.

The Road Ahead

We are going to see a massacre of “Middle-Tier” SaaS companies. The ones who refuse to let go of the subscription blanket will slowly suffocate. Their product roadmaps will become increasingly hostile to automation, trying to force humans to click buttons just to justify the monthly invoice.

They will become “Zombie Unicorns” -huge valuations, massive revenue, but functionally dead because they cannot adopt the technology that would save them.

“Only the paranoid survive.” — Andy Grove (Former CEO, Intel)

Meanwhile, the Vibecoders are coming.

They are building lightweight, usage-based tools that solve specific problems for fractions of a penny. They don’t have sales teams named Jason. They don’t have minimum seat counts. They just have an API key and a promise: We only make money when you do.

Let’s go back to that Zoom call.

After the CTO hung up on Jason, she didn’t just build a creative tool. She set up a usage-based billing account with OpenAI and an image generation API. She connected a few webhooks.

Her total cost for the month was $42.18.
The output was equivalent to three full-time designers.
Somewhere in a glass office in Manhattan, Jason is updating his sales forecast. He’s marking that deal as “Stalled / Nurture.” He thinks he just needs to send the CTO a better case study. He thinks if he just explains the value of the “Enterprise Synergy Dashboard” one more time, she will cave.

Jason doesn’t know it yet, but he’s selling buggy whips to a woman driving a Tesla. And the meter is running.

Originally published at https://www.amitgoel.me on December 4, 2025.

The Rate Card Is Dead: Why Charging Rent for ‘Features’ Is Financial Suicide in an Outcome Economy was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

Let’s start with a bedtime story that is currently keeping the General Counsel of a $10 billion company awake at night.

In June of this year (2025), a developer named Michael Luo (known online as “AzianMike”) decided he was tired of paying DocuSign $15 a month to sign three PDFs.1 He didn’t complain on X (formerly Twitter). He didn’t switch to a competitor.

He opened an IDE called Cursor, invoked a few AI agents (Claude 3.5 Sonnet, mostly), and “vibecoded” a tool called Inkless.

He didn’t write the code line-by-line. He didn’t hire a PM, a Designer, or a QA engineer. He just… vibed. He told the AI what he wanted: “Make me a drag-and-drop e-signature tool that sends a PDF to an email.”

Two days later, Inkless was live. It was free. It worked perfectly.

DocuSign’s response was not to innovate. It was to panic. They sent Luo a Cease and Desist letter so aggressive it could have been written by a T-800. They claimed “intellectual property infringement” and “false and misleading statements.”

Why? Because Michael Luo hadn’t just built a clone. He had exposed the dirty, trillion-dollar secret of the SaaS industry: Most “Enterprise Software” is just a CRUD database with a Stripe subscription attached.

“The reality of building web apps in 2025 is that it’s a bit like assembling IKEA furniture. You don’t make the wood. You just put it together.” — Andrej Karpathy (Founding member, OpenAI)

DocuSign isn’t a tech company; it’s a landlord. And the tenants just figured out they can build their own house for the price of a sandwich.

The “Golden Age” of Rent-Seeking (2010–2024)

To understand the murder, you have to look at the victim. For the last 15 years, we lived in the Golden Age of SaaS. The business model was beautiful in its simplicity:

1. Find a human friction point (Sales, HR, Inventory).

2. Build a workflow wrapper around a database.

3. Charge $50/user/month forever.

Investors loved it. We called it “Recurring Revenue.” It was predictable. It was safe. It minted billionaires. But it relied on a massive barrier to entry: Software was hard to build.

If you were a Digital Media Publisher, you couldn’t just “build” your own ad server or CRM. You had to buy Salesforce and Google Ad Manager. You were a hostage to their roadmap and their pricing.

The Content Publisher Who Walked Away

But that hostage situation is over. I recently witnessed the CTO of a mid-sized digital media publisher (let’s call them “DailyTech”) torch a $120,000/year contract that combined a legacy CRM and a complex Ad Server license.

For years, they paid this “AdTech Tax” because they believed the vendor’s hype that managing ad inventory required “proprietary algorithms” and “black box magic.” It didn’t.

Here is what happened:
The CTO took her 5 in-house engineers. They didn’t start a 12-month migration project. They spent three weeks “vibecoding.”

• The CRM: They prompted an AI to build a React front-end that mirrored exactly how their sales team booked campaigns. They stripped out the 50 unnecessary clicks the SaaS tool forced on them and built a streamlined interface that just did one thing: sold ads.

• The Ad Server: They used AI agents to write the ad-serving logic that connected directly to their content management system (CMS) and analytics.

• The Cost: A few thousand dollars in developer hours and API tokens.

The Result:
They now have a proprietary Ad Tech stack that fits their business like a glove. It is faster, it has zero license fees, and they own the IP. The SaaS vendor didn’t just lose a customer; they lost the ability to argue that their software was hard to build.

The “50 to 5” Engineering Shift

This aligns perfectly with what Sarah Guo, founder of Conviction VC, predicted:
“We are seeing a shift where clients are replacing 50-person engineering teams with 5-person ‘AI Architect’ teams. The bottleneck is no longer code; it is clarity of thought.”

Companies are no longer buying software; they are generating it. The “Make vs. Buy” calculation has flipped. “Make” used to be risky. Now, “Make” is instant, disposable, and perfectly customized.

The Economic Theory: Why Subscriptions Are Dead

This isn’t just a tech shift; it’s a macroeconomic correction. The SaaS model is built on Access Economics. You pay for access to the tool, usually via a “seat license.” But in an AI world, access to code is abundant.

Let’s look at the economic forces destroying the subscription model.

1. The Jevons Paradox of Code

Economists love the Jevons Paradox: As technology increases the efficiency with which a resource is used, the total consumption of that resource increases rather than decreases.

In software, this means: As the cost of writing code drops to zero, the amount of software in the world will explode.

We will have more software than ever before. But we will pay for less of it. We will have millions of “Disposable Apps”-software that exists for a single project, a single week, or a single day, and is then deleted. You don’t pay a monthly subscription for a disposable fork.

“If the cost of intelligence trends to zero, the value of ‘renting’ intelligence (SaaS) trends to zero. The value shifts to the outcome.” — Sam Altman (CEO, OpenAI)

2. The Death of Seat-Based Pricing

The subscription model assumes that value is tied to human headcount. The more people using the software, the more value you get. This is fundamentally broken in an AI world.

• Old World: You hire 10 SDRs. You buy 10 Salesforce licenses. Value = 10 humans working.

• New World: You hire 1 AI Sales Agent. It does the work of 100 SDRs. You buy 0 Salesforce licenses.

Why would a company pay for “seats” when the goal of AI is to reduce the number of seats needed? The economic incentive is now to eliminate the human user, which eliminates the SaaS revenue stream.

Porter’s Five Forces: The SaaS Massacre

If you apply Michael Porter’s famous strategy framework to a typical “Workflow SaaS” company (like a Project Management tool or a CRM) in 2025, it isn’t just a threat; it is an extinction-level event.

1. Threat of New Entrants (EXTREME)

• The Reality: The “Moat” of development cost has evaporated.

• The Impact: Two teenagers in a basement with an LLM can now replicate the core feature set of a $100M ARR company in a weekend. There is no longer a “minimum capital requirement” to start a software company. Every developer is a potential rival.

2. Threat of Substitutes (EXTREME)

• The Reality: The biggest competitor to Salesforce is no longer HubSpot; it is The Customer.

• The Impact: The “Make vs. Buy” decision has fundamentally shifted. When “Making” takes 3 days and costs $50, the substitute for buying SaaS is building a custom internal tool. The “Home Cooked Meal” is finally cheaper and better than the restaurant.

3. Bargaining Power of Buyers (INFINITE)

• The Reality: Vendor lock-in is dead.

• The Impact: Data portability is mandated by law (GDPR) and facilitated by AI agents that can map schemas instantly. If a SaaS vendor raises prices by 10%, the buyer can simply prompt an agent: “Export all my data from Asana, build me a clone in React, and import the data.” Churn is no longer painful; it’s trivial.

4. Bargaining Power of Suppliers (HIGH)

• The Reality: SaaS companies are no longer the source of intelligence; they are resellers.

• The Impact: The “Suppliers” are the Foundation Model providers (OpenAI, Anthropic, Google). SaaS companies are just UI wrappers around these models. The suppliers hold all the leverage. If OpenAI raises API costs or releases a feature that eats your wrapper (like “Canvas”), you are dead. You are a middleman in a world that hates middlemen.

5. Rivalry Among Existing Competitors (CUTTHROAT)

• The Reality: Feature Commoditization.

• The Impact: When everyone uses the same AI models to generate code and features, differentiation drops to zero. If you launch a cool new “AI Summary” feature, your competitor will vibecode the same feature by Tuesday. This leads to a brutal price war, driving margins down to the cost of compute.

Industry Deep Dive: The AdTech “Survival Matrix”

Let’s apply this to AdTech, an industry notorious for bloating workflows and charging a “tax” on every dollar. Using the lens of Vibecoding vs. Agents, we can categorize the industry into two buckets: The Dead (Commodities) and The Living (Outcomes).

1. The Commodities (Ripe for Vibecoding)

• Reporting Dashboards (Datorama, Tableau):

Why it dies: A dashboard is just a SQL query visualized. A Marketing Manager can now drop a CSV into a customized LLM and say, “Graph my ROAS by region.” They don’t need a $50k/year license for a tool that just shows them their own data.

• Tag Management Systems (GTM wrappers):

Why it dies: It’s just JavaScript injection. An in-house engineer can vibecode a custom container script in 20 minutes that is lighter, faster, and free.

• Creative Resizing Tools:

Why it dies: Tools that charge per-seat to crop images for different banner sizes are finished. This is a native feature of every generative AI model now. It’s a feature, not a company.

• Insertion Order (IO) Workflow Management:

Why it dies: This is just “sending emails with PDFs.” (See: The DocuSign story).

2. The High-Value Outcomes (Ripe for Agency-as-a-Service)

These are tools that make decisions and generate revenue. Clients will pay a premium for these.

• The AI Media Trader (Replacing the DSP UI):

The Pivot: Don’t sell me a Demand Side Platform (DSP) where I have to log in and adjust bids. Sell me an Agent that connects to the exchange, analyzes the bid stream in real-time, and buys the inventory that converts.

The Model: I won’t pay a platform fee (%). I will pay a “Performance Fee” based on the ROAS lift you achieve.

• Dynamic Creative Optimization (DCO) Agents:

The Pivot: Don’t sell me a tool to “manage” my banners. Sell me an Agent that generates 1,000 variations, tests them live, kills the losers, and scales the winners automatically.

The Value: The value isn’t the tool; it’s the decision to swap the headline from “Buy Now” to “Shop Today.”

• Fraud Detection & Prevention:

The Pivot: This is “Liability-as-a-Service.” I will pay you to ensure I don’t buy bot traffic. I am paying for the outcome of clean traffic, not the workflow of looking at logs.

The AdTech Lesson: If your product allows a human to do work, you are a commodity. If your product replaces the human to generate profit, you are an Agent.

The Psychology of Delegation (Why We Will Still Pay)

If vibecoding is so easy, why will anyone pay for software ever again? Why won’t every company just build everything themselves?

The answer lies in Behavioral Economics and the Psychology of Delegation. We will still pay, but we will pay for different reasons.

1. The “Cognitive Miser” Theory
Humans are evolutionarily wired to conserve energy. Decision-making burns calories.

• The Insight: Even if I can vibecode a travel booking app in 10 minutes, I don’t want to. I don’t want to maintain it. I don’t want to think about the APIs breaking.

• The Opportunity: We will pay for software that acts as a “Complexity Abstraction Layer.” We pay not because we can’t do it, but because we are lazy. The software that survives will be the software that requires Zero Interaction.

2. Liability-as-a-Service (The “Blame” Market)
This is the darkest and most true aspect of corporate psychology. Managers buy software to mitigate risk.

• The Insight: If I build an internal HR tool and it accidentally deletes payroll data, I get fired. If I buy Workday and it deletes payroll data, Workday gets sued.

• The Opportunity: Companies like SAP, Oracle, and DocuSign might survive not because of their code, but because they sell Liability Insurance. You are paying for a throat to choke. You cannot sue a vibecoded script you wrote yourself.

3. The Principal-Agent Problem Solved
In traditional business, the “Principal” (You) hires an “Agent” (Employee/Software) to do a task. Usually, the incentives are misaligned (the employee wants to do the least work for the most pay).

• The Insight: In the Outcome Economy, AI Agents are the perfect employees. They have no ego, no fatigue, and total alignment with the prompt.

• The Opportunity: We will pay a premium for “Agentic Trust.” We will pay for the software that has the best track record of not hallucinating. We aren’t paying for the workflow; we are paying for the reliability of the decision.

“The future of business is not about who has the best software. It’s about who has the best judgement. Software is cheap; judgement is expensive.” — Naval Ravikant

Conclusion: Evolve or Be Vibecoded

To the SaaS founders reading this: Stop selling forms. Stop selling “digital transformation.” Stop selling “efficiency.”

If your product requires a human to log in, stare at a screen, and click buttons to move data from Column A to Column B, you are a “FormWiz.” You are charging rent for a house that anyone can now build for free.

You must evolve into Agency-as-a-Service.

• Don’t sell me a tool to manage my ads. Sell me an agent that optimizes my ROAS.

• Don’t sell me a tool to write code. Sell me an agent that ships features.

As for DocuSign? They can send all the cease and desist letters they want. But they should remember what happened to the music industry when they tried to sue Napster. They didn’t stop digital music. They just made sure they weren’t the ones selling it.

Closing thought: The next billion-dollar software company won’t have 5,000 employees. It will have 10 employees and 50,000 agents. And they won’t be charging a subscription-they’ll be taking a cut of the profits.

Originally published at https://www.amitgoel.me on November 30, 2025.

The Rent Is Too Damn High: How Vibecoding and The Outcome Economy Are Killing the SaaS Dinosaurs was originally published in Amit ‘s Colliding Neurons on Medium, where people are continuing the conversation by highlighting and responding to this story.

It doesn’t matter whether you are a product manager or not, if you have ever tried building a product, there are three keywords that you would have come across for sure; Strategy, Roadmap and Prioritization. In all organizations, these three keywords are used in every context whenever there is ambiguity and decision making is hard. For every product manager, preparing for prioritization meeting is a nightmare. There is always a laundry list of features sent by different teams. Whether its the commercial teams or the account management or support teams, everyone has a list that is the highest priority blocking revenue and growth. There is not a single salesperson whose next major deal closure is NOT dependent on that one particular feature which you have been postponing for a long time.

In the last few years, I have come across so many product managers peddling so many frameworks. and on top of that, people ask for templates for those frameworks. There is RICE, JTBD, OKRs, Kano model, MoSCoW method, Opportunity Scoring and a lot more. These are just frameworks. There are hundreds of tools and templates based on all these frameworks interpreting their own way of building product strategy and roadmap to help product managers prioritize the features for development.

I have also read through all these frameworks. but interestingly, I found one thing hidden behind all these frameworks and that thing can be called as “common sense”. So, I thought of putting that common sense in the front. If you apply first principles, then there is always a base theory. With that base theory, a few people will develop certain frameworks, and then for the mass application, the rest of the world starts asking for templates. Ideally, if you can understand the base theory then you can develop your own framework without trying to fit your product needs into those hundreds of templates.

This article is my attempt to put that base “common-sense” theory upfront. But like all other product managers, I wanted to sound cool and knowledgable so I have given this “common sense” understanding a name calling it as “The GL(c) model”.

First thing first - The Theory ( a.k.a. Common Sense)

Every organization (lets consider for-profits product companies only in this article) exists to solve a large problem and in turn, generate revenue. and the aspiration of all these organizations is to have hockey-stick (exponential) growth over a period of time so that all shareholders can make money multi-fold. Its a rolling goal year or year for every organization. The revenue can be less in some years and some years can be magical with revenue growth beyond expectations. To achieve this revenue year on year, every product company builds a product strategy that needs to answer 3 basic questions.

1. What is that one single KPI that’ll change the odds in company’s favour ? ( Answer to this question results in product strategy )

2. To achieve that one single KPI, what are the key things that we need to work on ? ( Answer to this question results in product roadmap )

3. From those list of things to be done, what should be done first ? ( Answer to this question results in feature prioritization )

Now, if you look at all the frameworks, each of them is trying to answer either all or a combination of above questions. but as a product manager, when you learn to apply a framework without understanding the theory behind it, the outcome is a slide deck that you cannot defend easily. or the other outcome is a failed product.

The GL(c) Model

If you read the above questions again, you’ll able to guess the meaning of GL(c). Let me re-iterate it here.

Goal ( G ) - What is that one single KPI that’ll change the odds in company’s favour ?

Every product is built over years and decades. In the beginning, there is a vision and mission. Then, there is an MVP. Then you get to version 2 or 3 before getting to product market fit (PMF). Once you get the PMF, you need to scale. After the scale, you need to diversify and build multiple associated product lines. This cycle evolves over a number of years but no one can build a strategy that can last for even 5 years. The vision and mission can largely remain unchanged but what changes is the strategy. And the strategy is generally always is to execute “successfully” in the expected direction.

The “Goal” in the GL(c) model is the statement (or KPI) valid for a pre-defined period that can define “success”.

Let’s take the example of twitter (I still find it hard to call it X). What do you think can be the product goal of twitter in 2024 ? is it to grow the subscribers of premium plan or premium plus plan or will it be to grow the ad revenue ? Also, isn’t it contradictory to their own product strategy. if they want to grow premium subscribers, then their ad revenue will decline. and if ad revenue increases, then they can’t focus on getting premium subscribers. So, how do you think they should define their goal and build product strategy ?

The answer can be (and this is my imagination) that in all cases, they need user growth on their platform. For user growth, they need to drive a lot more engagement on their platform.

so, the fundamental goal can be “Drive Daily Active User growth to greater than 20% MoM by EOY “.

This goal is directly linked to total revenue without worrying about paid users or ad revenue. Once they define this as a goal for 2024 ( or for next 2 years), their product and business strategy will get defined across sub-product lines (subscriptions, feed algorithm, ads, engagement, SSO etc). Each product line can now build their own roadmap based on the goal.

Don’t worry, I’ll shortly talk about levers and contribution and it’ll all start making sense.

Levers ( L ) - To achieve that one single KPI, what are the key things that we need to work on ?

Levers can be simply defined as one or more features that are required to achieve the goal. This is the actual work done by the engineers as part of product development. But then, I am calling them levers and not features or epics. The difference is subtle but important. Feature or Epic is the actual definition of work to be done. but it becomes a lever when it has a KPI that moves the needle towards the goal. Lets say you are able to acquire 10 clients for your product through sales effort. Now, out of those 10 clients, only 2 actually spent USD 10k or above every month, rest of them actually spent only USD 2k or less in a quarter. So, by “definition of done” given in the epic, that epic is still complete as it is working and clients are using it. but is it really moving the needle towards the goal?

And that’s why levers are those features that have explicit one single KPI defined for themselves without being dependent on other factors outside of that feature. Ask a question while defining a feature: Will this feature on its own can deliver the expected value measured by the KPI you identified ? if yes, it is a lever. if not, its just a feature that might make a product look good but might not be contributing to the success of the product.

Again, let’s continue the twitter example: We have set a goal above. Lets build some imaginary levers. ( by the way, you can always group multiple levers to give the collection a logical name or define it as a sub-product like I mentioned in the goal).

Now that the goal is to drive active user growth by 20% MOM, we need to list a few development items across the board (don’t worry prioritization is the next step) on each sub-product.

If you look at the levers, each of the work item will have a very specific KPI that doesn’t not depend on any other feature. As a product manager, you’ll still need to write detailed specifications of each feature, but more than that, you need to define the KPI in such a way that it directly impacts the goal. Proxy metrics like Average Revenue per user, Average number of subscribers, Clicks on ads etc. try to create an impression that they affect the goal but actually they are psuedo metrics that help you justify your job, not the success of the product.

Contribution ( c ) - From those list of things to be done, what should be done first ?

Now that you have defined the goal and outlined the levers, how do you define what should be done first. One of the most important thing to understand is that levers don’t start working from day one. It takes time to see results. For example: in Jan 2024, when you’ll build you strategy and roadmap slide decks, you cannot expect the goal to be achieved in Feb or March 2024. Engineering needs time and effort. Then, the product hits the market. Clients start using it and then, you see the results. So, It can be even 6-9 months before you even see the first dot on the chart. Keeping that in mind, the formula of contribution is very simple. For each lever, check the impact (outcome) against the effort required. The idea of contribution is to do the things that require the lowest effort giving the maximum impact. Those are the best levers to be completed first. At the same time, things that need high effort also need to get kicked off in parallel and that’s why quantifying contribution of each lever in terms of percentage contribution towards the goal becomes all the more important. This contribution percentage will allow the teams to plan resource allocation, hiring, skill set mapping and all the other engineering and commercial activities like product marketing, commercial activations, partnership discussions etc.

Let’s continue the twitter example :

Disable Ads for premium plus subscribers.

Contribution : Let’s say we can gain 1 million users by end of year assuming DAU for premium plus is 50% of total premium plus users. Total twitter users are 100 million. Effort required is 3 engineers for 5 months. but Strategic value is “Very High” or 20 percent points. So, we get 20.5% contribution with an effort of 15 person months.

Reduce the number of ads to 1 per 8 items in the feed.

Contribution : Similar to above calculations (imaginary in this case, but some grounded calculations in real world), can help determine that contribution of this lever will be 30% with effort of just 10 person months

and similarly, for rest of 3 levers, contribution will be 10%, 15% and 40% with effort of 80, 20 and 15 person months respectively. Once the contribution is known, not only it helps in building the product prioritization chart, but it also helps in planning all company wide activities like marketing, sales, support, hiring, financials etc.

The Last Step

If you put this all together,

1. you can actually represent the product strategy and a prioritized roadmap in one single table that be embedded in a maximum of 1-2 slides.

2. You can easily say “No” to all feature requests that do not move the needle.

3. Across the organization, people can actually think through about their feature requests before reaching out to you.

Lastly, and the most important thing, Build a dashboard using all the KPIs associated with the levers and a chart for the goal. As the dashboard gets updated automatically, the strategic conversations become easy. A few things that can happen afterwards are

1. Strategy Evaluation : Are we moving in the right direction at the right speed ? Are we investing resources in the right place ?

2. Accountability : If all investments are as per strategy, is sales moving at the right pace ? are we pacing ahead or behind ? If we are behind, what are the levers that need to be changed?

3. Goal Audit : Did we pick a conservative goal or an ambitious goal ?

4. Market Analysis : if everything is great, what new markets we can enter or what new products we can innovate ? Can we change our product positioning ?

I strongly believe that if you follow this method and try to answer the 3 basic questions, you don’t need to learn any frameworks. Actually, believe it or not but first principles based approach works because it is just based on common sense and common sense doesn’t need a framework or template.

Seed references to know more

I strongly recommend the following 5 most important books to read if you want to understand the common sense behind product strategy, roadmap and prioritization.

1. Understanding Michael Porter by Joan Magretta

2. Thinking in Systems by Donella H. Meadows

3. The Goal by Eliyahu M. Goldratt

4. High Output Management by Andy Grove

5. The Subtle Art of not giving a Fck by Mark Manson

Actually the fifth book in the list has nothing to do with product planning. i just liked it and its a fun read.

Feel free to follow me on Twitter ( @amitgoel78 ) if you like to discuss about product management.

Disclaimer : I am using Twitter as just an example and I neither have connection with Twitter(X) nor I have any information about their plans. The usecases listed are only for illustrative purposes based on my limited understanding of their product. I cannot use the real world examples from my past experiences as it can reveal confidential information and I do not want to risk myself to litigation. The last thing I want is to give away my hard earned money to “Suits”.

In 2009, My team and I were working on a prototype assuming that one day TVs will be connected to the internet and when that happens, advertising will become programmatic similar to what was online display advertising on the internet in those days. but more than that, people will also have second screen devices and the content they’ll be watching on the TV will be influenced with their second screen devices activity. AWS cloud was catching up steam and getting popular and we were working on how to get STB software on the cloud and also, build a developer platform for TV apps just like android play store or apple app store.

During the project discussions, one of my team members started a pitch to management with an underlying assumption that it’ll not be TVs that’ll be connected but Set Top boxes will be connected to internet. While his experience if STB domain was limited, he had a rich experience of working with one of the largest TV OEMs in the world. He was quite confident that TV manufacturers will always depend on STB or similar devices for any content consumption. Also, this was due to the fact that the company we worked for was one of the largest STB software / Pay TV company in the world. He was very confident about his idea as it would have resonated well with the management too given that everyone wanted STB market to grow leaps and bounds. at that time, Connected TVs was an idea that could have killed the STB industry in itself. Netflix was a small player and was contained to desktops/mobile devices.

To build the pitch, he gathered all data points from his previous experience and research and along with STB industry reports, he started working on the case. During a few discussions over a month, couple of us asked him to think about STB market and the risks associated with it. Technically, enabling an STB with internet seemed a lot more challenging that enabling a TV with the internet. And applying the Porter’s five forces, if TV manufacturers were ready to invest in HD and 4K televisions, adding internet capability will be a peanut cost to a television compared to an STB and a threat of substitute was imminent too. As a manager, I still wanted to make sure that we had all the options open on the table. but my team member was adamant on his proposal.

A few weeks back, during recent discussion online with a group of product managers about blind spots that a product manager can face, I got reminded of this past episode. and that made me talk about Dunning-Kruger Effect to those product managers.

Dunning-Kruger Effect

For those who do not know what Dunning Kruger effect is :

The Dunning–Kruger effect is a cognitive bias whereby people with low ability, expertise, or experience regarding a certain type of a task or area of knowledge tend to overestimate their ability or knowledge.

In their research paper published in 1999, David Dunning and Justin Kruger implied that people who have limited experience are more confident about their knowledge compared to the people who are vastly experienced. This high confidence with limited knowledge stems from the ignorance they have about their own knowledge levels or capabilities resulting in them thinking that they know a lot more than others and hence, falsely makes them believe that they know it all. In reality, people with real experience and knowledge tend to have a room for self-doubt on their knowledge levels thinking that others in the same domain will also have similar knowledge levels and hence, there may be things to learn and possibly, there are more unknowns to discover.

In context of product management, this is highly relevant. Product managers deal will high level of ambiguity on a regular basis and are responsible for influencing decisions that can affect the future of organizations. Being a high visible and high stakes role of product manager, it is all the more required to be careful to evaluate all aspects of the discussion, being open to adverse feedback, listen to completely tangential viewpoints and then, try to present the probabilities of the various options rather than presenting one’s own line of thought as a fact.

It is well know saying “A players hire A players, B players hire C Players” . What it means is people who really understand their own capabilities want to work with similar high calibre people but people with mediocre capabilities want to hire less capable folks so that they can be confident and manage superiority over others. In most of the cases, B players are always have high confidence levels with quite less intellectual humility and it is visible even during interviews when they are asking questions to prove that they knew more than the candidate. This is nothing but a representation of Dunning-Kruger effect where low calibre people overestimate their ability and end up rejecting high calibre people because such smart people had different opinions or ideas that did not match the expectations.

Intellectual Humility

To achieve this, intellectual humility is highly important for a product manager. But people who face Dunning-Kruger Effect also lack in intellectual humility. Intellectual humility can be defined as the importance of knowing that one can be wrong. As a product manager, we learn management methodologies like SWOT ( Strength, Weakness, Opportunities, Threats), Porter’s five forces and many other theories either in business schools or by experience. but to actually apply these theories or methodologies, intellectual humility is a lot more important. Being humble helps us listen to the opportunities and challenges that others might know about. but if a person is facing Dunning-Kruger effect, it also makes him arrogant at the same time. it makes the person thinks that he knows everything but then the world is not listening to him and makes him go into a downward spiral where his or her team player abilities get affected without him realising about it.

Many of you might have heard of Amazon’s 14 leadership principles and how they apply these principles in their interviewing process. One of the amazon’s 14 leadership principle is “Are right .. a lot”. It is defined as “Leaders are right a lot. They have strong judgement and good instincts. They seek diverse perspectives and work to disconfirm their beliefs.”

This is a great principle for a product manager. To execute any ambigious open ended project, product managers need to be right, a lot. But that come with a caveat of intellectual humility. That’s what it means in the second part of that definition when it asks to seek diverse perspectives and work to disconfirm their beliefs. But for a person with Dunning-Kruger effect, that’s where confirmation bias steps in.

People with Dunning-Kruger Effect seek out diverse opinions or perspectives to depict themselves as a team player but in reality, it is to prove their point so that they can claim they are right, a lot. Confirmation bias drives people to identify only those opportunities or data points that help them to confirm their beliefs. To persevere with one’s beliefs, the person can even start interpreting the data that goes against the belief into a favourable condition or completely ignore that point. Confirmation bias makes people selective in their choice of data points. Sometimes, it makes them selective to an extent that they choose to work with people (or invite only those people in meetings) who will tend to agree with them or have similar opinion and ignore the rest who may have a different opinion.

Such situations are a hotbed for conflicts. This is a perfect situation for a PM behavioural interview question “tell me an example when you didn’t agree with one of your team mates and how you arrived at a decision”. The answer to this question will reveal certain traits of a person on dealing with conflicting situations but more than that, it may also reveal about how a person thinks about himself, whether he is highly confident about his own views but still considers others opinions seriously or not. Even if people practice to answer this question several times, the way they’ll answer will still give away a lot.

How to avoid such conflicts?

The idea of any discussion is to focus on the problem and not on the people. There are always exceptions to this rule where some discussions are done only to escalate a conflict (pun-intended towards the politicians across the world)

For a product manager, admitting mistakes or acknowledging the knowledge gaps help in reducing the potential errors and sets us up for success. Also, it helps in building trust about our capabilities in the teams we work with. As a PM, we should consider all the estimations and data published as probabilities and not facts. When some data or statement is considered a probability, our brain makes us think about its chances of failure, in turn, making us anxious. Anxiety forces us to think a lot more on edge case scenarios or “pit-falls” that can be avoided. This anxiety also makes us consult not only people who are subject matter experts but also others who can guide us out of this ambiguous problem statement helping us opening up our mind to broader set of possibilities. Thus, we become more upfront in outlining challenges and threats in decision making. That makes us a calculated risk taker while being proactive and optimist at the same time.

Intellectual humility is the path that every product manager should follow to be aware of one’s own knowledge gaps. Once we are aware of our own knowledge gaps, it makes us honest in our conversations and helps saying “I don’t know” a lot more than ever. Every time a product manager says “I don’t know”, it helps to go back, study, learn a lot more about the topic to gain knowledge and to become a real expert. Any expert we meet today will always add an element of doubt in their statements even if they are almost 100% sure about them being right. That element of doubt stems from their awareness and knowledge of probability of almost everything in the world can never be 100%. There is always a chance of failure and that chance makes them intellectually honest and humble.

Epilogue

The incident I mentioned in the beginning happened in 2009. As with all the jobs, people move on. I moved on in 2012 to become an entrepreneur. My team member moved on to join another company. That incident just became one of those vague work memories that don’t make any sense in hindsight. In 2013, Netflix launched the series “House of Cards” changing the world of content production and TV viewing. We are in 2022 now. While Set Top Boxes still exist, Connected TVs have taken over the world. Even the biggest of satellite TV providers moved to online streaming. The advertising models have changed due to CTV ( acronym for connected TV) being a dominant force. Not only the TVs became connected, the data became the new oil that feeds into advertising. What we called second screen devices then is just a context with respect to TV but essentially, personal devices are the ones that drove the change.

Seed references to know more

1. Wikipedia : Dunning-Kruger Effect

2. Paper : Dunning-Kruger Effect

3. Wikipedia : Confirmation Bias

4. Understanding Michael Porter

In 2001, I was working on developing a software licensing system that managed distributed license keys for any software using server based floating licensing mechanism. Yes, Remember the old days of client server architecture and buying software was still not so common as it was expensive. Nostalgia aside, I was a software engineer coding in Java who believed that the system was working fine and my piece of code was working as expected. So, I sent the code for review to my manager ( whom I “later on” started respecting a lot for his programming and analytical skills) who just took a brief look at the code and rejected the whole component within 30 minutes. It was my 20 days of effort that he didn’t care about. Of these 20 days, I had spent last two days continuously working in office with almost no sleep and surviving on pizza, coke and coffee.

I asked him the reasons and he explained that my basic assumption on one of the encryption mechanism was wrong and that I needed to rewrite the whole component. I went back and started working again. Just that, this time I tried to fix it by repurposing the code with a supposedly correct assumption. After a few days, my manager again rejected the code and explained me another issue I wasn’t able to see through. By the third time, I had lost my patience and I argued with my manager thinking that he doesn’t know what he is talking about.

The IKEA Effect

Have you ever faced a situation when you worked very hard on a product or a feature spending weeks and months and then, someone on your team started asking the basis of your fundamental assumptions ?

Most probably, yes. That happens with most of us. But the fun part is when the questions continue and you fail to understand why the team member is asking so many questions. It seems to you that the team member is either being stupid or doesn’t like you or the product you are working on. This is where the “IKEA effect” is taking place.

Prof. Dan Ariely described the “IKEA effect” in his 2011 paper that he published along with Prof. Michael Norton and Prof. Daniel Mochon as “labor alone can be sufficient to induce greater liking for the fruits of one’s labor: even constructing a standardized bureau, an arduous, solitary task, can lead people to overvalue their (often poorly constructed) creations.”

As a product manager, we spend so much effort in building the product requirements and work on achieving the product market fit trying to reduce failure as much as possible. But as we do that, we keep on moving on an iterative path where our effort ( measured as sunk cost) is getting bigger as the days go by. With the sunk cost getting higher, we get into a defensive justification mode whenever some one asks us a seemingly harmless or basic question. This is for a simple reason that we start over valuing the effort we have put in such that we dread to change the path ( if required) as it may lead to delayed timelines or pseudo loss of reputation or supposedly , may get us bad performance rating ( if that ever happens).

The Endowment Effect, Loss Aversion and Status Quo

Now, as we invest more time and effort, our sunk cost is getting higher making us attached emotionally to the product we are working on. That leads us to another problem called “endowment effect”.

Prof. Daniel Kahnemann, along with Prof. Jack Knetsch and Prof. Richard Thaler explained the endowment effect in his 1991 paper describing it as people tend to value the items they own more highly than they would if those items did not belong to them. If you have invested your time and effort heavily on the product that you are working on, as a product manager, it is natural human tendency to develop an affinity towards the product. Sometimes, it becomes so difficult to separate your own identity from the product. Identify the scenarios when some product managers start calling a product as their baby because they conceptualised the idea and have worked on it day and night.

Both “IKEA Effect” and “Endowment Effect” stem from a basic cognitive bias called “Loss Aversion”. Loss Aversion is described as the fear of loss is greater than the joy of gain.

Assume that the product you are working on is moving very smoothly and is on target for release. While, everything is going good, a colleague on another team ask you some questions about the way feature works and she suggests a change which will result in a change of direction for your product but at the same time, may result in 50% gain in estimated revenue. but the impact of making that change is delay in project release by 6 months.

While we all believe that we think rationally as we are rational people, most of the product managers will still fear the loss and hence, try to avoid the loss by continuing on the path as the fear of losing 6 months can be considered greater than 50% gain in estimated revenue. if the company work culture allows for taking a short term hit in favour of long term benefit, the product manager needs to change the direction but that may result in heartburns in the various teams and significant stakeholder management. So, a risk averse product manager may rationalise the decision of sticking to the course and not change direction, citing Sunk Cost, indirectly justifying the decision and maintaining the status quo.

When this status quo is maintained and the questions get postponed so as to find the answers at later stage, this results in inability of finding the product market fit for many products and then, products pivot at a late stage leading to losses mounting to millions of dollars in so many cases.

Whether it is “IKEA Effect”, “Endowment Effect”, “Loss Aversion” or “Status Quo Bias”, it doesn’t matter if you don’t remember these terms. What matters most is the way these unconscious biases can be avoided by training your subconscious mind to look out for the symptoms. The way you train your mind is by following a simple common sense technique called “First Principle Thinking”.

First Principle Thinking

“First Principle Thinking” helps you to ask the very fundamentals questions that form the premise of the problem you are trying to solve. Whether it is “Socratic Questioning” method or “5 Whys” of the famous author Simon Sinek, or a child like inquisitive method of asking questions, just go down to base of the problem and why you are really interested in solving the problem. Identify the win-win formula for everyone and do not stop asking the “Why” until you are convinced that there is no further doubt or question that can arise. if a question arises somewhere along the path even after you started, go back to the drawing board for course correction and cut your losses early. This is where when someone said “Fail Fast & Fail Often”.

Long story short: watch out for the feedback signals you receive, even a faintest signal can help you fix the problems early on and recognise the gaps in your plan.

By the way, I did go back to my manager who rejected my code multiple times after a few weeks to understand his explanation once again and that’s when, before explaining the software design, the first thing he taught me was the concept of “loss aversion” and a varied form of “first principle thinking”. That helped me see beyond the personal biases and also, learnt for the first time about something called “cognitive biases”.

References

1. The “IKEA Effect” : When labor leads to love

2. Anomalies : The Endowment Effect, Loss Aversion and Status Quo Bias

3. First Principles : The building blocks of true knowledge

4. The Feynman Learning Technique

5. Thinking Fast and Slow

Products are built the same way life has evolved and will keep on evolving in the future. The Red Queen Effect defines it well that how organisms must constantly adapt, evolve and proliferate in order to survive when pitted against ever evolving rival organisms in a constantly changing environment.

Most of the technology companies across the world have very similar organisational structures. They follow the same standard rules of small teams like Amazon’s 2 pizza rule, Spotify’s Squad concept or any other name given by companies to this structure of having small but supposedly autonomous teams. Generally, These teams are a set of engineers, product manager, designer, testers etc. These small teams own individual parts of the product or the product itself and then, come together to execute the vision of the company defined “collectively” by the leadership of the organization.

Image Courtesy : https://medium.com/productmanagement101/spotify-squad-framework-part-i-8f74bcfcd761

Any product that has the right mix of vision, alignment and autonomy at every level has a significant chance to be successful. As a cultural practice, the senior management always declares that the teams are autonomous and provides them with vision and alignment. Alignment with the vision of the company can fuel the autonomy to go in the right direction.

I assume that your organizational structure is similar to the one defined above following best practices of agile methodology, short release cycles, constantly measuring performance of product and people, social harmony in the teams and yet, you know that the project or the product feature never gets released on time or much worse, product releases do not match the expectations of the customers, making company lose both money and reputation.

Once the crisis starts building up (which is a regular thing in most companies known as “always on” firefighting mode), the senior management tries to take control of the situation. But then, you’ll notice that this is a repeated behaviour in most of the organisations resulting in frustrations, demotivated employees, a big difference in performance of people and with 20% of people driving the rest of the 80% people in the organization to achieve the targets until all hell break loose.

And this crisis generally happens due to two behavioural issues : Optimism Bias and Strategic Misrepresentation

Optimism Bias

Optimism bias is defined as a cognitive bias causing someone to believe that they themselves are less likely to experience a negative event. It is also known as unrealistic optimism or comparative optimism.

It’s a standard statement for managers that whatever the software developer estimates, multiply it by 3 and then, the sales guys multiply that figure by 3 again to arrive at a timeline that can be given to customers. But still, many times, projects fail to deliver on time with expected quality and feature set.

There are multiple factors that contribute to over optimism. Most of the time, senior management is driven by sales figures. To close an account, promises have to be made. And it is a well known fact that salespersons are expected to be short-sighted and to live their life quarter by quarter, For a salesperson, only the current quarter matters because their incentives are dependent on quarterly results. This percolates to the top management including the CEO whose incentives are tied to revenue performance and market growth of the company. Only a handful of CEOs have the vision and ability to let go of short term small gains for long term enormous benefits. Like an individual, most of the CEOs believe that the money in hand is much better than the promise of money in future.

As per NAO ( National Audit Office) UK report “Over Optimism in government projects”, there are five factors that contribute to over-optimism. To understand the details, Please read the NAO report here.

When these factors of over optimism come into play, culture of the organization plays an extensive role in identifying the optimism bias and enabling the teams for the right decision making by allowing to use the autonomy given to them. But then, if the alignment shakes a bit on Y Axis (as given above in alignment vs. autonomy diagram), results change and the output of the product suffers.

Strategic Misrepresentation

In the case of strategic misrepresentation, planners and promoters deliberately misrepresent costs, benefits, and risks in order to increase the likelihood that it is their projects, and not the competition’s, that gain approval and funding.

While the definition of strategic misrepresentation says that management teams of an organization does it deliberately and is always intentional, inherently it can also happen because of optimism bias. Every organisation wants to win every contract and acquire every new customer to increase average revenue per customer and their market share. In a bid to achieve its goals and to build the company image alongside, facts start to get misrepresented (intentionally or due to optimism bias) during discussions with potential and existing customers.

Generally, Misrepresentation means lying or fraud. But it can be also be loosely defined as over-stating your organization’s or your product’s capabilities. It also can be about promising certain product features set that don’t exist as of today but the organization is leveraging the time taken by customers in decision making. Typically, a misrepresentation goes unnoticed if the customer gets the desired product. But in most of the cases, that doesn’t happen and it leads to friction between the organization and the customer creating undue pressure on the squads or teams to deliver higher quality product much earlier than expected disrupting the planning and product roadmap and in turn, setting the butterfly effect in motion. Refer to this article for understanding butterfly effect in context of product management. Why Good Products Fail Despite Having Best Engineering Teams And Process Management Systems ?

You are what you measure : Dan Ariely

Human beings adjust behavior based on the metrics they’re held against. Anything you measure will impel a person to optimize his score on that metric. What you measure is what you’ll get. Period.

Dan Ariely, a renowned author and professor of Psychology at Duke University wrote this in his article ( You are what you measure ) published in HBS in June 2010.

Essentially, its the Key Performance Indicators (KPIs) that govern the behaviour of an individual ( whether in life or in an organization). If the CEO is incentivised like salespersons to achieve higher targets on revenue, he or she will build plans in such a way that other factors of a product like technical superiority, quality, experience will take a back seat. It will happen because earning revenue early in the year will help achieve his or her targets faster. As Dan Ariely mentioned in the article, If the CEO is allocated high number of stocks as renumeration and also has a stock price target given by the board, his or her decision making process will somehow give undue advantage to stock price such as to increase stock price at any cost. This decision of driving stock price up will change the alignment and vision of the organization which may not match with product capabilities resulting in directional shift of the product.

As depicted in the above given diagram, when organizational pressures are too high, not only the management, even development teams tend to be less autonomous and less truthful in turn, end up misrepresenting and under estimating project complexity resulting in shorter timelines and half baked product features. On the contrary, if the organizational pressures are too low, the teams across all levels suffer from optimism bias, over-estimating their capabilities in project execution.

In their book “Thinking Fast and Slow”, Nobel Laureate Daniel Kahnemann & Amos Tversky identified this phenomenon for the first time and termed it as “The Planning Fallacy”. And defined this behaviour of people as “Inside View”.

The Planning Fallacy, using the concept of optimism bias, proved the errors of judgment are not random but are systemic and predictable. Most of the CEOs ( and in general human beings) fall prey to the planning fallacy because of the way their incentives are structured. So, when organisations are looking for successive victories every week, every month, every quarter looking for short term wins, the planning fallacy manifests in the behaviours of the senior management of the organization and then, sets up the organization culture for the same in some way or the other.

Kahnemann and Tversky suggested “Outside View” as a solution to avoid both optimism bias and “unintentional” strategic misrepresentation.“Outside View” is also known as “Reference class Forecasting”. Outside View mechanism suggests to consider the planning factors and stages of similar products or projects considering similar resources. It also suggests to consider empirical data in the same reference class of project under consideration. Building a reference class of similar products or projects and studying their details ( identifying success and failure data points) and then, creating a baseline can help in reducing the optimism bias and also, help the management teams in reducing the misrepresentation of facts about the product. This definitely helps in identification of mis-selling practices by sales teams who tend to over promise to close deals faster and to achieve their quarterly targets.

Having said that, “Outside View” suggests not to over simplify the problem at hand but to take a macro view of the impact the project will have from the perspective of customers and market on the organization as a whole.

In the end, as done in every sprint cycle following Agile methodology, conducting a retrospective or a post-mortem for every project helps in identifying the biases that were at play during the execution of the project. This helps in identifying the “invisible gorilla” or “elephant in the room” from the perspective of organizational culture, its project execution philosophy and its processes built to make the company successful.

References :

1. https://www.nao.org.uk/wp-content/uploads/2013/12/10320-001-Over-optimism-in-government-projects.pdf

2. https://en.wikipedia.org/wiki/Optimism_bias

3. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.307.5094&rep=rep1&type=pdf

4. https://hbr.org/2010/06/column-you-are-what-you-measure

5. https://en.wikipedia.org/wiki/Red_Queen_hypothesis

6. https://medium.com/productmanagement101/spotify-squad-framework-part-i-8f74bcfcd761

7. https://en.wikipedia.org/wiki/Dan_Ariely

8. https://hbr.org/2012/08/the-planning-fallacy-and-the-i

9. https://www.amazon.com/Thinking-Fast-Slow-Daniel-Kahneman/dp/0374533555

10. https://www.researchgate.net/publication/237077814_THESIS_FOR_THE_DEGREE_OF_LICENTIATE_OF_ENGINEERING_Organising_the_Early_Design_Phase_in_a_Large_Infrastructure_Project

Did you know India’s per capita income for FY 2018-19 was Rs. 1,26,406 and for FY 2019-20, it is expected to be Rs. 1,35,048 ?

Per capita income can be defined as average income earned per person in a given area (city, region, country, etc.) in a specified year. It is calculated by dividing the area’s total income by its total population. It includes all people including children, elders or anyone who is counted in the census or population of the specified region.

That means “on an average”, every Indian ( including a child) earned approximately Rs. 1.35 Lakhs ( USD 1800) per annum last year. If that’s the case, “on an average”, the monthly income of an Indian would have been Rs. 11,254 ( USD 150) per month. Now, with that in mind, an individual would have earned Rs 375 per day “on an average”.

Since 2007, India has set its official threshold at Rs. 26 a day (USD 0.43) in rural areas and about Rs. 32 per day ( USD 0.53 ) in urban areas where people earning below these thresholds are defined as people living below poverty line (BPL) . As per RBI reports, Out of the total population living in the rural parts of 35 states and UTs of India, 25.7% of them is living below the poverty line. Isn’t it contradictory to say that on one hand, average income of an Indian Rs 375 per day where as on the other hand, 25.7% of Indians still live below poverty line ?

While the facts remain, the only thing that is making all of this look either good or bad is the law of averages. When you say “on an average” , you mean to say “typically”. What you may want to imply is that “typically” or “in general”, people are earning Rs 375/- per day based on the numbers available. But this does not paint the correct picture. In reality, “typically” does not mean average, it means “median”.

Outliers are single biggest influencers while calculating average of any collection of data points. Now, if you are an analyst graduated from a business school who has studied analytics as a subject, you can apply normal distribution curve, calculate distance from mean to get standard deviation. And then, arrive at many other conclusions. The only thing standard deviation can help with is to understand that if the distance from the mean is too big or not. Standard Deviation can help in understanding whether considering the average as a metric is useful or not.

Considering the above, According to the 2015 World Wealth Report, India had 198,000 high net worth individuals (annual income over USD 1 million) with a combined wealth of USD 785 billion. Assuming India’s population to be 1.3 billion people, 0.015% people generate more than half of total income of Indian population. This is what causes India’s per capita income to reach such a high number despite 25.7% population being below the poverty line.

In calculating averages, sample size matters a lot. Consider this as an example. In a room, there are 10 people with an average net worth of Rs 75 lakhs ( USD 100k ) . Now, enters Mukesh Ambani in the room whose net worth is Rs. 3.75 lakh crores ( USD 53 billion) . Suddenly, the average net worth in the room increased from USD 100k to USD 4.8 billion.

Unbelievable? Isn’t It a mathematical fallacy? You must be thinking it is crazy to put Mukesh Ambani in the same room to calculate average networth of middle class people. But that’s what we do everyday in our jobs. We calculate averages all the time. If you are salesperson, you always have to beat the average sales numbers without consideration of equality in the sales figures and other factors.

I am not sure of other industries as I have spent 19 years working in this so called IT Industry or Tech industry in the tech hub of India called Bangalore. Every year, when the time comes for appraisals and increments, HR heads of every company brings data to the table proving how much “on an average” the increments are being handed out this year. In that average calculation, HRs very smartly make sure to remove Flipkart, Amazon, Google, Facebook like companies but make sure to include the likes of Infosys, Wipro, TCS, CTS etc. and then, they provide the salary increment guidelines to the senior management to have 1-2% above the average. This number is then made public giving a false sense that the company is paying above average salaries or increments.

As I have been working in media technology company for last few years, I have noticed a very cool trend of publishing average viewership numbers. As per the latest report published by BARC ( the company that measures TRPs for Indian TV channels), due to COVID lockdown in India, the average TV viewership duration has increased to 4.48 hours per day from average viewership of 3.46 hours per day. This may ignore a convenient fact that most of the times, people leave the TV on and no one is watching. And secondly, the sample size of BARC is just 1,88,000 people.

So, if you notice the numbers in these examples and may be the examples you’ll see in your life, averages are calculated and used only when they convey a positive indicator and create a false sense of pride and growth. Most of the organisations have been using averages for the purpose of marketing and managing stakeholders. But for real business planning and strategies, average is the number that should be avoided all the time. If you building a business strategy based on law of averages, I am sure you are just trying to raise money from VCs or trying to prove that the work you have done is great.

What should you do if averages should not be used ?

There are so many mathematical ways to determine the success or failure of the business. And one thing is for sure, to determine business health, depending on simple math is dangerous. But for the sake of simplicity, one of the better indicator is Median. Median is the number that lies in the middle of the dataset. Outliers do not have any impact on Median. Median gives you just a bit but somewhat better picture of the data you have. As it ignores outliers, you can safely assume that if median number looks good for your business, then you are doing good “typically”.

Otherwise, the better way is always to create clusters of datasets. Histograms can help you create these packets of data or various smaller datasets. Now, by segmenting the data, you’ll be able to create buckets of the origins of data and then, perform separate business analysis of each bucket. Once you do that, you can drive a strategy either to improve each bucket performance, or to kill some buckets and focus on buckets that are doing better.

Consider an example from the media industry itself. People have started watching a lot of content online. Apart from YouTube as a popular content source, Netflix, Amazon prime video, Hotstar, Sony Liv, Zee5, Voot, MXPlayer, and plethora of other apps are getting to the market. Each of the apps are claiming that they are winning a huge market share as they are growing. Now, if you are working in one of those media companies, instead of looking at average session duration of a viewer or average viewership duration in a day, look at the median viewership duration. This will give you a realistic view of how much time “typically” people spend on your platform. Then, create the bucket of users based on different session durations. By doing that, you’ll be able to profile the interests of the users on your platform. May be your platform is more famous for short form content rather than movies or may be its popular only for Cricket matches or only for episodic content.

While there are so many statistical analysis techniques that are applied in every business, the basic technique of calculating an average is too common at every level in any professional environment. Because of the law of averages ( and subsequently either law of small numbers or law of large numbers), management teams arrive at quick conclusions that can affect livelihoods and earnings of people and their own business goals too.

In a nutshell, If the average number looks good, call your marketing department to make press releases and announcements on social media. Otherwise, ignore any kind of averages in life.

No one has ever liked to be just an average in life.

References :

1. http://statisticstimes.com/economy/gdp-capita-of-india.php

2. https://www.indiatoday.in/india/north/story/planning-commission-bpl-earn-rs-25-a-day-india-141619-2011-09-21

3. http://mospi.nic.in/sites/default/files/publication_reports/India_in_figures-2018_rev.pdf

4. https://www.financialexpress.com/economy/around-22-indians-live-below-poverty-line-chattisgarh-jharkhand-fare-worst/1713365/

5. https://en.wikipedia.org/wiki/Mukesh_Ambani

6. https://en.wikipedia.org/wiki/Histogram

7. https://www.livemint.com/industry/media/tv-records-762-million-viewers-per-week-in-2019-despite-nto-barc-study-11584735703190.html

8. https://economictimes.indiatimes.com/industry/media/entertainment/media/india-records-highest-tv-consumption-in-a-week-with-total-time-spent-of-1-27-trillion-minutes/articleshow/75069638.cms?from=mdr

9. https://www.investopedia.com/terms/i/income-per-capita.asp

10. https://en.wikipedia.org/wiki/Income_inequality_in_India

11. https://www.cia.gov/library/publications/the-world-factbook/

Almost a year back, I met some of my old-time friends ( going back 20 years or so) for lunch. It was like a get together of 40+ year old men who were meeting after two long decades to catch up like bachelors. The meeting started with lunch and went on to dinner late in the night. The topic of discussion varied from how we all started our life at the same work place when we were just in our early 20s and never imagined that we can make money enough to afford luxurious restaurants and fancy cars with multiple trips abroad. We made fun of each other on how we struggled, borrowed the money from each and never paid back till date. The amounts in discussion were laughable in today’s time ( like Rs 300 or so ) but in those times, it became a matter of survival. The fun continued for long…

But by dinner, the discussion became a bit serious. The same people who were having fun discussing the past, now became serious discussing the future. Out of nowhere, one guy started talking about retirement. And lo and behold, here were five 40+ men who had no clue about “retirement”. Retirement as a concept was always in our minds but it never occurred to us that we had crossed our equator (or half life) of our professional life. All of us knew that people retire and saw our parents retiring from work. We all saved money but never saved the money thinking that we’ll stop working one day either by choice or by force.

So, One of my friends started the discussion saying that to live a decent life with moderate lifestyle, Rs. 1,00,000 a month is enough. Then, the debate started on the definition of “decent lifestyle”. What was “good enough” for one was “not good enough” for another. Even the basic amount of currently monthly expense varied for each of us. While the discussion started heating up, another friend of mine said that it is good enough to have a corpus of INR 1 crore ( INR 10 million).

And suddenly, every one turned into an expert economist talking about terms like GDP, inflation, rising school fees, rising fuel prices, and at the same time, rising income etc. I was one of those expert economists too who didn’t even know the service tax we pay on the food that we were eating. But then, in a group of five 40+ year old men, who dares to say that I don’t know shit about economy. Anyway, the discussion progressed, everyone left after dinner and soon we forgot about our retirement discussions.

Now, with the Corona Virus spread across globe, I’ve ended up working from home for more than a month. With this COVID19 discussions on WhatsApp groups and online forums, there is another discussion happening, and that’s about RECESSION and UNEMPLOYMENT. I have been reading news about people getting fired from jobs across the world ( including India). Salaries are being cut significantly. People are being furloughed or asked to go on unpaid leaves. Supply chain is in trouble. Demand is falling down due to people staying home and are in lockdown situations.

While all this is temporary and being an optimist, I believe that things will be back on track sooner than later, It made me think about my future. And I thought what will happen If I lose my job or I am not able to work. While working at home, I noticed that I was working much harder and much longer hours that I worked while being in office. I started thinking that If I don’t feel like working one day and want to follow the FIRE ( Financially Independent, Retire Early) theory, how much money do I need in my account so that I don’t have to go back to work or live below the level that I live currently.

I started thinking about it and applied some of my excel skills ( along with some basic 10th standard maths) to calculate what should be my retirement corpus. And to achieve that, what should be the monthly amount of money I should start saving to build a savings portfolio and achieve my goals of a peaceful retired life ( following the FIRE philosophy)

So, After a few hours of work, I came up with this simple excel sheet that helped me visualise my future requirement of money.

Retirement Corpus Calculator : Click here to download the Microsoft Excel File

for example : If I have a monthly expense of INR 1 Lakh ( INR 1,00,000) , then after 15 years when I retire, my monthly expense will be more than INR 2 Lakh ( INR 2,07,893)

and then, I can see some of the following calcutions for generating a pattern of my savings and my expenses after retirement.

And the results were astonishing. I was laughing at myself thinking that the amount that looks so big today can be a meagre amount in a decade or so. For example : 10 years ago, Rs. 100/- could buy you 10 litres of milk or 5 loafs of bread. Today, the same Rs. 100/- gets you only 2.5 litres or less of milk and may be 2 or 3 loafs of bread. It seems, prices have more than doubled in 10 years or so.

So, I was laughing at myself thinking how foolish I was not to notice the compounding effect of inflation on the prices in years while being an “expert armchair economist” discussing the GDP and economic policies of India.

We, generally, don’t notice it today. But think that if this recession stays longer than required and has the same impact like the “great depression” during the 1930s, our earnings will go for a toss. Leave the increments aside, we may not have enough for current lifestyle too and that can set us back by few years of savings, in turn, making us work longer than required number of years to achieve our goals.

While it may seem simple to you today, it is not that straightforward to plan for retirement. Most of the Indian IT workforce has never seen retirement and the first generation of IT industry in India is getting into retirement zone in this new decade. So, the Indians who caused the IT revolution and started a trend of high income or high salaries along with foreign travels are yet to see the effect of retirement and did they save enough or not.

I don’t have much to say, but wanted to make people aware of the impact of economy and recession and how things can possibly play out. It’s better to plan and have a visibility into what lies ahead. Based on this, you should be able to plan your investment strategies, insurances, mediclaim policies etc.

Here is the link again : Retirement Corpus Calculator : Click here to download the Microsoft Excel File

Let me know what you think of this calculator. Like and share this article on social media (LinkedIn, Facebook, Twitter or anything else that you use) if you think this calculator really helped you visualise your retirement and help you calculate the corpus you may need.

“It used to be thought that the events that changed the world were things like big bombs, maniac politicians, huge earthquakes, or vast population movements, but it has now been realized that this is a very old-fashioned view held by people totally out of touch with modern thought. The things that change the world, according to Chaos theory, are the tiny things. A butterfly flaps its wings in the Amazonian jungle, and subsequently a storm ravages half of Europe.” — from Good Omens, by Terry Pratchett and Neil Gaiman

With the corona virus pandemic growing exponentially across the world, we are all locked down in our homes. At the time of writing this article, I have been at home without venturing out for 15 days and I still have at least 15 more days to go like this. While more than 30,000 people are dead with more than 6,00,000 infected, the pandemic is still growing with no signs of slowing down.

Dr. Li Wenliang from China had discovered the Corona virus strain in early December. He notified his colleagues and wanted to alert about the SARS kind of virus that he had found out. He wanted to alert the world that something deadly is coming up. But chinese authorities shut him down before he could say much. To the extent that he was made to apologise publicly for rumour mongering. While the chinese government was successful in shutting him down, they failed to notice the spread of corona virus. By the time they realised the scale of this disease, it was too late and few cities were completely taken down. Within a month, the virus had spread to South East Asian countries and parts of Europe. European countries failed to notice it again causing a pandemic and now, the USA is leading with most number of people infected with corona virus among all countries in the world. This pandemic not only caused the issue of healthcare around the world but has also caused the early onset of economic slump or recession at an unprecedented scale that will result in mass scale unemployment, loss of trade, extreme supply with low demand, real estate price crash, healthcare crisis with social and emotional instability. And it will leave the world scarred for the next 3-5 years making this event as one of those that are written in history books being mentioned for next 100 years at least.

What Dr. Wenliang had noticed or discovered was the initial change in the ecosystem that caused the butterfly effect to take place.

What is Butterfly Effect ?

As given in wikipedia: In chaos theory, the butterfly effect is the sensitive dependence on initial conditions in which a small change in one state of a deterministic nonlinear system can result in large differences in a later state. In a simple language, butterfly effect can be characterised as a small event triggering a chain of events having huge implications on a system such that the system can go into an irrecoverable state. Butterfly effect is something that almost everyone fails to notice because the tiny event that causes it may not even be related to the large catastrophic events that it can cause.

What is the connection of butterfly effect and good products ?

Although the concept applies to all kinds of products, I’ll focus on B2B software / tech products. I have spent my last 20 years building B2B tech products in different roles and have been part of many products being successful or going to the grave. All B2B products companies always have focus on some kind of verticals as they develop products related to particular industry or domain. For example : GE healthcare sells its products to hospitals, Slack works mostly with tech companies (not so much with Banks or manufacturing sector), Stripe works with companies who need to collect online payments, Akamai works closely with media industry, and so on…

All these companies have product managers focussing on the product growth and strategy while working with engineering to define the “right” product. And the senior management is looking at the top line and bottom line while keeping sharp eye on competition and market share. But then, many good products or companies die even after having a successful run. Why ?

Let’s consider a hypothetical example :

A good B2B product from company ABC was being bought by many large customers in the market. ABC was growing as customers are adopting the product fast. As the product was growing, ABC was setting up processes and departments to help customers use the product effectively. Be it 24/7 support, project and account management, finance support, product support, engineering interfacing or anything like that, ABC was making sure that the customer is able to reach out for any help. but even after all this, one small customer (say D) was complaining about certain problems in the product that was causing her some stress. The stress was not only financial but also emotional as she was losing face in her own organization. all the departments of ABC while trying hard were still struggling to resolve the problem. ABC’s management had taken notice but it seemed like an isolated incident as other customers are running fine.

Over a period of 1 year, ABC added 3 new customers in its portfolio but lost about 8 existing customers and in the following year, ABC wasn’t able to sustain the product and had to shut down.

What had happened in these two years?

In the B2B world, while the industries are large with huge market sizes, people are socially connected and talk to each other. All of the industries ( be it airlines, media, healthcare, banking, etc.) have their conferences and have their offices in business hubs in different cities. People meet and talk. This small customer (D) started discussing its problem about ABC’s product with another customer ( E ) of ABC who, in general, was happy. But when the second customer (E) heard about the problem , she also started thinking a bit about it and questioned about the effectiveness of the product in her own organization. Over a period of time, vendors and competitors got the wind of the issues and the word got out. While the company ABC was trying to figure out the issues with the product, they started losing customers faster than they were winning deals. Sales team of ABC was facing the heat in the market. Existing and potential customers were losing trust on ABC and its product. Sales cycle had become much longer than earlier. This lead to frustration causing distrust internally in different departments and in turn, leading to blame game. The system fell like a house of cards.

That small customer’s complaint was the initial change that had caused the butterfly effect. It started a chain of events which led to other customers losing trust in the product and identify the issues that weren’t even noticed earlier. When this happened, it created a lot of pressure on engineering and product teams to roll out features and fix issues faster , in turn, causing more issues and more frustration for the customer. This chain of events caused a downward spiral when the competition kicked in and customers of ABC started the switchover.

Finally, a good product having potential of leading the market disappeared from the scene “overnight”.

What customers buy ? Looking through the lens of Jobs To Be Done framework

While the product is same, every customer buy a product with different expectations. While someone may buy an SUV for long trips with family, another person may buy an SUV to drive it as a taxi and earn livelihood. SUV is still the same. Media companies use CDN to distribute content across the world, but some want to keep the costs low but the others may want the best streaming experience. CDN will still be the same.

So, how does the same product cater to different expectations when “one size fits all” doesn’t work as a strategy at least in B2B environment ? To achieve this, companies create processes. But then most of the companies that fail create processes to support the product, not the customer. While the support department may be called customer support, it is actually supporting and defending the product by making fixes to keep the product running and to make the customer complaint go away. Sales is pitching the product to customers, not to solve her problem but to sell its product and earn incentives. Engineering is writing code to make the feature work without worrying about scale and performance. Management is just reviewing the margin numbers.

Clayton M. Christensen says, “People buy products and services to get a job done”.

The late Harvard Professor and a well known author innovated a concept or framework called “Jobs To Be Done”.

The theory behind this framework is to identify why the customer is buying the product. In B2B domain, while an organization is paying another organisation for the product, its humans who are decision makers and influencers. Even in B2B deals, product features are an important consideration but the buyer is also putting her own reputation at stake inside her organization when she recommends any product. This just makes it more complex as buyer has locked in her emotional side into the product even in this B2B decision. When the product doesn’t perform “as expected”, while the support department of buyer is interacting with support department of seller, there is a feeling of discontent where people start questioning the decision makers and influencers. In order of save themselves, the buyer starts discussing issues in social forums connecting with other buyers seeding doubt in the minds of others too. When the network effect happens, It doesn’t consider the positive or negative side of effect. It can go either way and generally, negative trending happens much faster than the positive one.

Ideally, while building products and processes, the companies should consider the nature of the jobs that the customer wants to accomplish. Everyone knows about your product and its features so you don’t need to project it. During business discussions, discover the answer to a simple question “Why is this customer buying our product and what problem is she trying to solve ? “

Then, just categorise that one answer into different categories of jobs that are not just functional but social and emotional. This means the buyer has certain unique context and circumstances that doesn’t apply to any other buyer.

Once you identify the job, create a job statement that has 3 key components : Action, Object , Context. A simple template of a job statement can be :

Once you create the main job, create some related jobs and prioritise them by asking the buyer to rate the jobs in order of priority. This helps understand the needs of the buyer. Once that happens, now you can align your internal company processes to the needs of the buyer. While every organization has the same kind of departments or org structures, the smooth functioning of the departments depends on how they interact with each other with respect to each customer. For one customer, functional product might be a priority leading to more communication with product support and engineering, but for another financial impact might be more important than features of the product leading to proactive communication with analysts of the company. And may be for the third customer, features or financial impact is lower priority than the impact it may have on the end user experience.

These different job outcomes involve the same product, may be 90% of the same features and same teams with same processes but the priority of the jobs and customer need drives the requirements of where the emphasis should be in the process pipeline and can help in identify what you think is okay can be a bottleneck for the customer.

Product and Processes are important but if they do not provide a smooth experience and does not help in accomplishing the job for the customer then they don’t matter. When we think process, we think assembly line. But we always forget that there are always multiple paths in the same assembly line to address different needs of different customers.

If the companies apply “Jobs To Be Done” framework, they may still not notice those small events having the capability of causing butterfly effects but can definitely reduce the probability of those small events significantly and protect their products from failure.

By the way, two months later in early February 2020, Dr. Li Wenliang died from the very same disease that he discovered.

“A very small cause which escapes our notice determines a considerable effect that we cannot fail to see, and then we say the effect is due to chance. If we knew exactly the laws of nature and the situation of the universe at the initial moment, we could predict exactly the situation of that same universe at a succeeding moment. But even if it were the case that the natural laws had no longer any secret for us, we could still only know the initial situation approximately. If that enabled us to predict the succeeding situation with the same approximation, that is all we require, and we should say that the phenomenon had been predicted, that it is governed by laws. But it is not always so; it may happen that small differences in the initial conditions produce very great ones in the final phenomena. A small error in the former will produce an enormous error in the latter. Prediction becomes impossible, and we have the fortuitous phenomenon.” — Jules Henri Poincaré (1854–1912)

References & Suggested Reading:

1. Deterministic Nonperiodic Flow - Edward N. Lorenz

2. The Butterfly Effect: Everything You Need to Know About This Powerful Mental Model

3. The Butterfly Effect: Managing Your Organization as a System

4. Butterfly Effect

5. Know Your Customers’ “Jobs to Be Done”

6. Competing Against Luck: The Story of Innovation and Customer Choice - Book by Clayton M. Christensen

7. Dr. Li Wenliang

Before I start expressing myself in this highly opinionated article about “Ed-tech” landscape of this country and the hype created by these companies, let me give you some numbers that are published by MHRD, the ministry responsible for the Indian education system.

As per MHRD report of 2016 (yes, that’s when they counted last for enrolment in Indian education system), India has more than 260 million students. and as of MHRD-AISHE report of 2019, there are 37.4 million students studying for higher education ( under-graduate and above). If you go into the details of AISHE report, following are some key points standing out in the summary :

• About 79.8% of the students are enrolled in Undergraduate level programme. 1,69,170 students are enrolled in Ph.D. that is less than 0.5% of the total student enrolment.

• Maximum numbers of Students are enrolled in B.A. programme followed by B.Sc. and B.Com. programmes. 10 Programmes out of approximately 187 cover 80.3% of the total students enrolled in higher education.

• At Ph.D. level, maximum number of students are enrolled in Science stream followed by Engineering and Technology. On the other hand at Post Graduate level maximum students are enrolled in Social Science stream and Management comes at number two.

• Distance enrolment constitutes about 10.62% of the total enrolment in higher education, of which 44.15% are female students. 10.62% means approximately 3.8 million students are enrolled in educational institutes through distance study mechanism.

• There are 993 Universities, 39931 Colleges and 10725 Stand Alone Institutions listed on AISHE web portal and out of them 962 Universities, 38179 Colleges and 9190 Stand Alone Institutions have responded during the survey. 298 Universities are affiliating i.e. having Colleges.

• 16.3% of the Colleges are having enrolment less than 100 and only 4% Colleges have enrolment more than 3000.

• Share of female students is lowest in Institutions of National Importance followed by State Private Open Universities, Deemed Universities-Government.

If you read these numbers, the least you can understand is the enormity of the system and how complex it must be given many factors of India demography like :

• India is a home to more than 122 languages and of which23 are considered as official languages of the country.

• Indian population is 1.3 billion of which 65% is under 35 years of age.

• Unemployment rate of India is at a high of 8.4% with urban unemployment at 9.6% and rural unemployment at 7.8%

• Indian GDP is at 5% down from 6.8% in 2018 and 8.2% in 2016.

• as per census 2011, 60% of Indian population lives with an income of Rupees 200/- per day or less.

• Approximately 1.6 million engineers graduates in India every year of which only 10% or less are employment ready.

Now, that you know the basic data points let’s begin the storyline.

The Opening Scene

Indian education system has long been struggling to produce people that are skilled to generate enough income for a better quality of life or a standard of living. In the pre-liberalization era, Indians were looking for a job security. With the license raj in place and government jobs only providing a regular source of income, the dependency on skilled workforce wasn’t that high. But in 1991, when the economic system changed and India became open to world market driven by divestments and free market dynamics, a lot of private sector companies started emerging.

But by late 90s along with the private sector boom, job security started going down while incomes started rising. This started the creation of income divide where some people started earning a lot more than most of the others. Along with that, came the technology boom where India became the back office of the world and software services industry took a leap in the future by offering skilled but cheap labour from the US market price point. This drove the growth of Infosys, WIPRO, TCS, and the likes of them. When Infosys offered a starting salary of Rupees 15,000/- per month against an average income of Rupees 5,000/- to 8,000/- per month of a non-IT sector employee, the face of Indian education system started changing. Before early 90s, people applied of engineering when they were really interested in it and a lot of people wanted to be lawyers, architects, farmers, shopkeepers or may be, government employees. but by late 90s, Indian middle class aspired to be a software engineer. This drove the culture of getting into engineering college and even, if you joined in civil engineering , you were still picked up by the likes of Infosys and TCS , trained on the job for mainframe computing while just fixing the Y2K errors and making a killing of 15K-20K rupees a month with a few months of US travel.

Suddenly, the influx of engineering colleges happened and from a less than 100 engineering colleges, India had more than 5000 engineering institutes in a span of few years. This grew to such an extent that as of today, if you go for a drive on Indian highways, you may not find food , water or a toilet but you’ll definitely find an engineering college in middle of nowhere. This sudden requirement of being the back office of the world or the cheap version of silicon valley drove the aspirational index of parents and students in India. The rush of being rich, foreign travel, peer pressure of owning a car and a house drove people to ignore what they were really good at and pursue engineering at any cost.

This mad rush towards engineering ( and a minor rush towards medicine as a career) gave birth to a new supplemental education market- The coaching industry. The tuition centres / coaching classes mushroomed in all cities and towns helping students prepare for the entrance exams of engineering and medicine. While a student paid a substantial amount to schools as a fees, 10X amount was being spent on the coaching classes. Thus, the Kota Factory came into existence and suddenly, parents were spending lakhs of rupees ( in many cases, mortgaging their jewellery and houses) to get their children admitted to these coaching classes. As per news reports, many of the teachers who provide IIT entrance exam coaching in the town of Kota earn annual salary of multiples of crores which is way more than most of their students will ever earn.

While this is going on in the country on one side, the other side of the country is struggling to get its GDP on track, providing enough for its citizens by reducing unemployment and bridge the ever widening rich and poor divide. For that, education system needs to change towards skilling of people in various occupations like lab technicians, nurses, clerks, bank officers, salespersons, masons, architects etc. But while there is demand for such skilled people, there is no supply as everyone wants to be an engineer.

Indian Ed-Tech Landscape

In the midst of all this, India is seeing another boom and this time, it’s the rush towards starting up. The success of Flipkart and Redbus has encouraged people (read: engineers) to become entrepreneurs. While this is a good sign, VC funding rush has driven to an extent that the founders are raising money, and then burning money so fast that they are vanishing at a faster rate than they should be. Among all the kinds of startups, “Ed-Tech” ( Education Technology) has also become a buzz word. In a short period of time, entrepreneurs in the education world ( or as they are called “edupreneurs” ) have started launching online versions of the offline world and hence, the name “Ed-Tech”. If we have to classify Indian ed-tech startup scene, it can be done very quickly in following four segments.

Online Assessment : In old days, Brilliant Tutorials or Bansal classes used to have mock exam series and a booklet that was couriered to home for having practice tests. The same got translated in the online world and suddenly, you’ll find many companies offering exam prep services online. Essentially, a large enough question bank that students are playing around with. and 80% of the questions are same on all websites with almost 50% of the them being taken from last 20 years of printed books or test papers🙂

Online Coaching / Tutoring : A teacher sitting in front of the camera talking to students either in live video format or it is a recoded video that students can watch.

Online Career Counselling : Instead of going to a counselling at a CBSE accredited counselling centre, you can just pay a fee and have a chat or video call with a counsellor.

Executive Education Services : A mix of online / offline coaching and degree/certificate vendors for working professionals ( mostly engineering and management professionals).

Then there is Byju’s (“father of everyone”) doing everything mentioned above in one single platform and not only for senior secondary students but for kids who have just started their schooling in grade 1 or above.🙂

Before you take me wrong, Let me clarify this. I am not against any of them. It’s actually good that these companies have leveraged mobile apps and internet bandwidth to reach out of potential students and help them prepare better and get good marks such that students don’t need to get stuck and waste time in traffic but study at home at a cheaper cost. All Good till now. But then, why are these companies called Ed-Tech? They are not tech enough as they are not building cutting edge technology ( not even compared to e-commerce companies). And also, they are not education companies as most of them do not provide education or skill people but just help people get better marks and clear exams.

In my opinion, none of these companies are solving the real world problem of Indian education system. But they are just feeding on fears of parents and students. Students are rushing towards these online platforms because they do not want to miss anything that can cause them to score less. While the genesis of any for-profit company is to maximise revenue for itself, the companies operating in the sector of education and healthcare have also to bear responsibility towards the society. The advertising industry or e-commerce industry can consider only profits and margins but the education companies need to consider the fate of the student in its balance sheets. The quality of teachers on the Indian ed-tech companies cannot be measured as one does not know if they even know how to teach.

While at school and college level, it was still understood why these companies are leveraging the fear of people for profitability, In the executive education space too, the same is happening. Degrees and certificates from unknown / less known universities are being sold online along with placement offerings in the garb of blended learning.

US Ed-Tech Landscape

If you compare the US Ed-tech landscape, just do a quick google of the term Ed tech startups in US or read this article Education companies taking learning to the next level . Now, you’ll get the difference. While there are online coaching platforms and test prep companies in US too, there are a lot more others who are trying to solve the backbone of education system. There are companies who are building products to make schools and colleges function efficiently so that teachers can focus on teaching and not administration. There is companies providing a digital library, helping teachers build teaching material, predictive algorithm companies guiding students, fund raising companies for students, loan facilitators, learning management systems, building devices to improve learning experiences, and of course, companies that are focussed on skilling people in non-traditional jobs.

Yes. I know. US is a developed economy. They are an evolved country while India is an emerging economy and a growing market. But in an emerging economy, there is a lot more need to strengthen the backbone first. There are enough “Kendriya Vidyalayas” or “Public Schools” who can streamline their admission systems so that RTE ( Right to Education ) Act can be implemented effectively but then, there have to be technology companies building softwares to streamline processes. MHRD mandates that every school needs to have a counsellor and with that, also comes the responsibility that counsellors are doing their jobs effectively. There can be startups who can solve the problem of students needing a helpline where data is captured and shared with school to help the children. and the list of ideas for improving and making a difference in Indian education system can go on …

But then, Indian ed-tech startups are still stuck with entrance exams and coaching business. and even there , its the game of numbers and funding rather than making a real difference in life of children.

Byjus versus Khan Academy

Khan academy was launched in 2008 and had become hugely popular in the US in a very short duration. In the sector of K-12 education, Sal Khan, founder of Khan academy saw an opportunity and launched his venture as a non-profit. On similar lines, Byju’s saw the opportunity in the same domain of K-12 education in India and launched his for profit company Think and Learn in 2011. I have mentioned for profit and non-profit just to compare the motivations of the founders. There is no judgement on the same. In the end, if the children benefit for both of them, It really doesn’t matter if someone becomes a billionaire.

Byjus became famous very quickly, less for its courses, but more for its funding raised. Over the last 6 years, Byjus raised over USD 975 million from various investors. On the other hand, Khan academy being a non-profit, has raised grants or donations on an average of USD $30 million per year. In the year 2018, Khan academy managed donations and grants to the tune of USD 32.5 million. Compared to massive funding of Byjus, Khan academy is nowhere in terms on bankrolling its operations.

While Khan academy is completely free for the students, it still gets revenue of about USD 50 million on a yearly basis and expenses of about USD 45 million leaving them as well managed cash positive organization. On the other hand, Byjus being a private company doesn’t publish annual report but from media report, it seems they make a revenue of USD 200 million annually versus expenses of atleast USD 200 million or more. While Byjus claims they are a profitable company without giving out numbers, sponsoring Indian cricket team doesn’t come cheap for sure.

The only thing I am trying to drive a point while discussing the financial numbers is that while Khan academy is may be 1/5th or 1/10th of Byjus in terms of financials, the impact of Khan academy seems to be 10 times of Byjus.

While Byjus claimes to have 35 million registered users with about 2.4 million paying users and with a highly questionable practice of sign ups and renewals ( do read this :https://the-ken.com/story/the-loan-crisis-at-byjus/ and https://the-ken.com/blog/byjus-and-the-art-of-the-deal/ ) , Khan academy reaches to 71 million registered users for free.

Now, registered or paid users is not a metric for success of an education platform as it is very clear from the competitive nature of kids and their parents that they sign up for every platform that comes up in the market. The metric which should be looked at is the usage of the platform. Khan academy publishes that their platform is used atleast for 120 minutes a month but 1.5 million users and that’s what they consider as a success criteria. There were nearly 8 billion minutes of platform usage reported on Khan Academy for the year 2018. I couldn’t find the numbers of Byjus as Byjus doesn’t publish their platform usage statistics but while there is enough positive feedback about Khan academy, I am yet to come across posts on social media ( school forums, parents forums, references etc) where people have shown positive impact of Byjus.

That said, the only case it shows is while Byjus founder has become a billionaire by leveraging the standard FOMO of indian parents, Khan academy is the one who is still making a difference in the educational life of Indian kids while spending 10 times less than Byjus. While Khan academy is now referred by schools in US, Indian schools have actually started warning parents against the use of Byjus platform.

And at the same time, scale of Byjus funding and the modus operandi of sales had driven more Indian ed-tech startups to copy the model rather than really focus on solving the problem they actually started with.

Conclusion : What is the right thing ?

Honest answer. here is no conclusion. I don’t know. But I do know one thing is while making money is good and nothing evil about it, making money in education and healthcare without giving back the benefits what people deserve is no less than evil. Education and Healthcare are the only two sectors where people do not look for discounts, albeit, are willing to pay more to get the highest quality service. But this is one area where the startups are not raising the bar.

India startup ecosystem / entrepreneurs have huge opportunities in front of them to make a difference in the life of children and provide better education & skilling and make them ready to face the ever growing competitiveness in the world. By the time, Indian government will wake up to fix the system , entrepreneurs can be the one providing the support to the schools and universities while making money on the way. but then, in this mad rush of raising money from VCs and over-valued companies ( maybe Byjus is the next WeWork too alongside OYO), somewhere the mission and vision statements of these ed-tech companies is just an introductory slide on their pitch decks.

This article has too many topics to talk about that every section and sub section deserves many pages of its own with details and data points. But In this article, I have tried to cover the macro perspective of the whole education system in India so as to make you think harder. Humans, by nature, are competitive and if you have read Sapiens by Yuval Noah Hahari, you’ll understand it a lot more. So, blaming parents for competing and comparing is not so good. When the Indian population is growing and will beat China one day, the jobs and economy are not growing like China. on the contrary, jobs are shifting towards non-core sectors while ed-tech startups are still trying to mint money from the traditional models of coaching and exam preps.

The real ed-tech boom in India will happen when there is enough demand created in sectors other than IT & Banking, with the supply rationalised to make sure everyone makes a decent living . Till then, its just VCs throwing money at either IIT/IIM graduates who have no clue about teaching or education system or throwing money at companies just because some Zuckerberg has invested in them to get a share of Indian market. In both the cases, Indian student is just a number on the pitch deck of these companies with an ARPU ( Average Revenue Per User) shown with hockey stick curves. But in reality, the Indian students are still getting their supplementary education from the likes of Khan Academy or a Coursera.

References :

1. Economy of India - Wikipedia

2. Institutions of National Importance

3. Unemployment rate of India

4. Languages of India - Wikipedia

5. AISHE Report 2018-19 from MHRD, India

6. Educational Statistics at a glance - Report 2018 from MHRD, India

7. Statistical Year Book 2017 by Ministry of Statistics, India

8. Education in India - WENR

9. Engineering Education in India - Short and Medium Term perspectives by AICTE , India

10. Salary of Teachers in Kota

11. Khan Academy Financial Information

12. Byjus Cricket Sponsorship

13. Byjus grows revenue

14. Khan Academy - Leveling the playing field

BJP (Bharatiya Janata Party) got an astounding victory with majority (282 seats) in Lok Sabha 2014 elections (Parliamentary Elections )of India. This was an unprecedented victory of any political party in India in the last 30 years or so. Indian Parliamentary elections (also known as Lok Sabha elections) has 543 seats that go for public voting. Among the total seats, any party has to secure at least 272 seats (either on its own or by forming alliance with other parties) to win the elections. The winner gets the chance to form the government and rule the country for next 5 year term (at most).

I have tried to analyse the voting results conducted in 543 constituencies across the country during the Lok Sabha 2014 elections. I did it to assess or derive whether BJP will be able to repeat the voting patterns of 2014 and win the 2019 elections comfortably by securing the majority (at least 272 seats).

To achieve the objectives, this article analyses the data to derive the following :

1. Voter turnout ratio to find voting patterns across states in India which is demonstrated as consistent.

2. Percentage of vote share obtained by BJP to its voting margins in each constituency to demonstrate all states are not voting equally for BJP.

3. Normalise the BJP’s vote margin to get a normal distribution curve and to find the outlier states on either side to find the guaranteed wins and losses for BJP.

4. With various confidence levels, this article tries to find the confidence interval where it can be shown if BJP is able to win the majority or not.

Unfortunately, it seems BJP will struggle to get the majority and it will not be a smooth ride for BJP in 2019 Lok Sabha Elections.

This article considers the data only from the vote margins perspective for each constituency. The dataset considered has been explained in the next section. The input dataset and the resultant calculations can be downloaded from the below given links. elections-2019.xlsx

The complete Data set : https://www.dropbox.com/s/thp0c15wvxoh5fz/elections-2019.zip?dl=0

There are many other factors that go in predicting elections but are not considered in this article. For example : BJP’s performance being in government since 2014, country’s economic conditions, political climate, state elections wins and losses, public sentiment, demonetisation etc.

Explaining the Dataset

The data has been obtained from the Election Commission of India website and the link for the same is http://eci.nic.in/eci_main1/ElectionStatistics.aspx .

The master dataset contains the voting data from 2014 Lok Sabha Elections. This is the total population data for 508 constituencies out of 543 constituencies. This data was cross matched against 8355 records of all valid nominations of candidates in each constituency to calculate winning candidate and his/her margin along with the next trailing candidate along with vote share. Also, the data considers the number of total electors, total voters, gender based votes, postal votes, BJP’s winning and losing margins etc.

In this analysis, I considered the vote share of BJP as 0 (Zero) if BJP did not field any candidate from particular constituency and similarly, the vote margin is marked as -100%. The negative voting margin indicates BJP lost the seat while positive voting margin indicates the win. But to calculate normal distribution curve of BJP’s voting margins, the seats where BJP’s vote share is 0% is not considered as they did not field any candidate in that constituency.

The Voter Turnout

The average voting turnout in Lok Sabha 2014 Elections was 71.26% with an standard deviation of 9.32%. The median voting turnout was 71.42%.

Following graph indicates the voter turnout in each state and also, outlines the male voters, female voter turnout state-wise. This graph also outlines the voter turnout in each state as a percentage of national voter turnout. This helps in identifying the states that make an impact on the election results in a significant manner.

On closely analysing this graph, it figures that states like Uttar Pradesh, Maharashtra, Karnataka, Gujarat, Andhra Pradesh, Bihar, Odisha, Rajasthan, Tamil Nadu and West Bengal play a significant role in deciding the election results as they form the highest percentage of Voter turnout with respect to national voter turnout.

Out of these significant states, West Bengal and Tamil Nadu are not at all aligned to BJP. Even Andhra Pradesh is not so much favourable to BJP (whether it’s pre-poll or post poll alliance). Karnataka also has shown the tough fight given to BJP by JD(S) and Congress alliance in 2018 state elections.

Thus, It is in interest of any political party (whether BJP or any other party) to gain a significant number of seats in these states as voting in these states decide the fate of the party due to its sheer volume of vote share.

BJP Vote Share to BJP Vote Margin

To find the BJP’s strength to win elections in 2019 with majority, it is important to understand their vote share and winning margins in 2014 elections where they performed very well and went on to win 282 seats in total.

As we consider only 508 constituencies in this article, BJP won 268 out of the analysed 508 seats. There were 21 seats BJP won with less than 5% margin. This means there were at least 21 seats where BJP struggled but managed to win. And these 21 seats will provide a tough fight again in 2019 elections. Thus, BJP should give a special focus on these 21 seats to achieve the majority.

The above graph proves the point BJP had a very close contest in many of the constituencies where vote share and margin is almost similar. The huge negative difference between vote share and margin depicts that these seats are the losing seats and the gap is too huge to fill and should not be focussed upon as the voters may not be favouring BJP under any circumstances.

Finding the outliers

I decided to find the outlier states where BJP is guaranteed to win or lose. To do the same, I normalised the BJP’s winning margin percentage in each constituency by calculating its Z score.

To calculate the Z Score, I applied the formula

Z Score = ( Value — Mean ) / Standard Deviation.

Using the Z Score, I derived the probability density value (value for normal distribution) to plot the following normal distribution curve.

Also, I calculated the interquartile range (IQR) of BJP’s winning margin percentage. The upper fence of the same is 95.88% which means the highest winning margin in a particular constituency. And the lower fence is -57.33% which is the lowest margin where BJP lost the seat.

By applying 2 Standard deviations (each standard deviation is 40.06%), there are 56 losing outliers and 0 winning outliers. This means that there are 56 constituencies where BJP has lost elections with significant margin (beyond 2 standard deviations) and to win these, there is immense focus required. Ideally, winning these seats is practically impossible, BJP should write off these seats from their projection of majority wins. These 56 seats seem to be a guaranteed loss. Also, there seems to be 0 (zero) seats that provide a winning outlier or a guaranteed win. Hence, BJP should focus a lot more on seats with <=1 standard deviation and still a little less on seats > 1 standard deviation as they have still some comfortable position on >1 S.D. seats.

Will BJP be able to retain the majority (272 seats) in 2019 Lok Sabha Elections too ?

To arrive at a decision, I formulated the following NULL hypothesis

Ho = 272 ; Ho = BJP will be able to retain majority (272 seats) in 2019 Lok Sabha elections.

And the alternate hypothesis ,

Ha ≠ 272; Ha = BJP will NOT be able to retain majority (272 seats) in 2019 Lok Sabha elections

And I performed a two tailed test by using the Z value from the normal distribution curve. Also, I used the various confidence levels to find the confidence interval of number of winning seats.

In all possibilities of various confidence levels ranging from 99% to 50%, BJP still seems to win a minimum of 260 seats to a maximum of 266 seats.

Hence, I rejected the NULL hypothesis that BJP can retain the majority (272 seats) in Lok sabha 2019 Elections and accepted the alternate hypothesis.

Conclusion

With this analysis, I can safely derive the assumptions that BJP need to put very significant effort in securing wins even in the constituencies they feel safe and have won in the past. This is due to the fact that out of 282 secured wins in 2014 Lok Sabha elections, there are no guaranteed wins and there are 21 very less margin wins. But out of total 396 constituencies where BJP nominates its candidates, there are guaranteed 56 losses.

So, 2019 Lok Sabha elections is not going to be a cake walk and it seems BJP is going to struggle for majority.

To achieve majority, BJP needs to focus on identified low margin wins and increase their winning margin to safe region of within in 2 standard deviation.

References & Research Calculations :

1. Election Commission of India Website : http://eci.nic.in/eci_main1/ElectionStatistics.aspx

2. All the gathered datasets and worksheets : https://www.dropbox.com/s/thp0c15wvxoh5fz/elections-2019.zip?dl=0

3. Resultant Calculation Workbook : https://www.dropbox.com/s/oal30p88jl4d6jl/elections-2019.xlsx?dl=0