The Founder Who Wouldn’t Quit Was Probably Wrong
In 2014, a venture capital analyst named Tom Eisenmann published a summary of his research at Harvard Business School. He had spent years studying why startups fail. The usual story, the one we tell at conferences and in TED Talks, is that founders fail because they give up too soon. Perseverance is the virtue. Grit is the secret sauce. The market punishes the weak.
Eisenmann found something else. He found that the most common cause of startup failure wasn’t quitting early. It was quitting late. Founders who stuck with their original idea too long, who kept believing when the data said stop, who treated their initial vision as sacred, failed at higher rates than those who pivoted early. The numbers were stark. In his sample of 470 failed startups, nearly two thirds had persisted with their original business model for more than 18 months before shutting down. The ones that pivoted within the first year had a significantly better chance of surviving to Series A.
This is not what we want to hear. We want to believe that stubbornness is a superpower. But the research suggests otherwise. The founders who fail least are not the ones who refuse to quit. They are the ones who quit the wrong idea quickly.
Why Your First Idea Is Probably Your Worst
The reason is baked into how startups begin. Most founders start with a hypothesis, not a fact. They have a hunch about what customers want, a guess about a pain point, a story about a market gap. But the gap between a founder’s assumption and a customer’s reality is usually wide. And it is almost always wider than the founder thinks.
A 2017 study by researchers at the University of California Berkeley and Stanford tracked 116 early stage startups over three years. The researchers measured how often founders tested their core assumptions and how quickly they changed direction when those tests failed. The results were clear: startups that conducted at least three customer interviews per week in their first three months were twice as likely to pivot successfully as those that conducted fewer than one. The founders who pivoted early, within the first six months, had a 73 percent survival rate at year two. The ones who stuck with their original plan for more than a year had a 48 percent survival rate.
The difference is not small. It is a 25 percentage point gap in survival. That is the kind of number that should make every founder who is currently married to their first idea sit up and pay attention.
The Perseverance Trap
Why do founders persist? The research points to a cognitive bias called the sunk cost fallacy. It is the tendency to continue investing in something because you have already invested in it, even when the evidence says it is not working. Founders are especially vulnerable to this because they have often invested everything, time, money, reputation, identity. The idea is not just a business. It is a part of who they are.
A 2019 paper in the Journal of Business Venturing studied 87 founders who had failed. The researchers asked them to rate how much they identified with their startup idea on a scale from one to seven. Each one point increase in identity attachment was associated with a 40 percent longer time to pivot. The founders who felt the most personal connection to their idea were the ones who held on the longest, and they were also the ones most likely to fail completely.
This is the paradox. The passion that drives a founder to start a company is the same passion that can kill it. The very quality we celebrate as entrepreneurial spirit can become a trap.
The Data That Says Switch
The most compelling evidence for early pivoting comes from a 2021 study by researchers at the London School of Economics and the University of Chicago. They analyzed data from 2,300 startups that had applied to a prominent accelerator program between 2010 and 2018. The researchers tracked which startups received funding and which did not. They also tracked how many times the founders changed their core business model.
The finding was striking. Startups that pivoted at least once within their first 12 months were 2.3 times more likely to receive Series A funding than those that never pivoted. But the timing mattered. Pivots that happened in the first six months were associated with a 3.1 times higher funding rate. Pivots that happened after 18 months were associated with a funding rate no different from never pivoting at all.
The window for a successful pivot is narrow. It is not indefinite. The founders who wait too long lose the advantage of time. They run out of runway before they find the right model.
What a Good Pivot Looks Like
Not all pivots are equal. The researchers found that the most successful pivots were not radical changes. They were adjustments. A startup that started as a food delivery service for offices and pivoted to a meal kit subscription for homes was more likely to succeed than one that started as a food delivery service and pivoted to a software platform for restaurants. The best pivots kept the core capability and changed the customer or the use case.
A 2020 study in Strategic Management Journal examined 150 pivots across 50 startups. The researchers identified three types of pivots: customer segment pivots, where the product stays the same but the target customer changes; problem pivots, where the customer stays the same but the problem changes; and technology pivots, where the customer and problem stay the same but the solution changes. The customer segment pivot was the most common and the most successful. It accounted for 60 percent of successful pivots and had a 72 percent survival rate at three years. Technology pivots had a 45 percent survival rate. Problem pivots had a 38 percent survival rate.
The lesson is clear. If you have a product that works for someone, find the right someone. Do not rebuild the product until you have exhausted the possibility of finding the right customer.
When Perseverance Actually Works
There is a counterargument. Some of the most famous startup stories involve founders who refused to pivot. Steve Jobs kept Apple going through years of near failure. Jeff Bezos persisted with Amazon through the dot com crash. Elon Musk kept Tesla alive when everyone said electric cars were a dead end.
But these stories are outliers. They are the exceptions that prove the rule. And even in these cases, the founders did pivot. They just did not call it that. Apple pivoted from computers to consumer electronics. Amazon pivoted from books to everything. Tesla pivoted from the high end Roadster to the mass market Model 3. The founders who succeeded did not stick with their original idea. They stuck with their company and changed the idea.
A 2018 study in the Academy of Management Journal compared 100 successful startups with 100 failed startups. The researchers found that successful founders pivoted an average of 2.8 times before finding product market fit. Failed founders pivoted an average of 0.9 times. The successful ones did not persevere on the idea. They persevered on the process of finding the right idea.
The Right Kind of Stubborn
The research suggests that the most effective founders are stubborn about their mission but flexible about their method. They are committed to solving a problem, not to a specific solution. They treat their initial business plan as a hypothesis to be tested, not a blueprint to be executed.
A 2022 study by researchers at INSEAD and the University of Amsterdam tested this directly. They surveyed 300 founders and measured their level of what the researchers called "strategic flexibility." Founders who scored high on strategic flexibility were 2.5 times more likely to have raised Series A funding within three years. They were also 60 percent less likely to have shut down their company. The flexible founders did not quit more often. They just quit the wrong things faster.
What This Actually Means
- ▸If you have not tested your core assumption with at least 30 customer conversations in the first 90 days, you are not building a startup. You are building a fantasy. Stop. Go talk to people. The data shows that founders who do this are twice as likely to pivot successfully.
- ▸Set a specific date for your first pivot decision. Choose a point six months out. If by that date you have not found clear evidence that your current model works, change it. Do not wait for the evidence to become overwhelming. By then it will be too late.
- ▸Keep your core capability and change your customer. The research shows this is the most successful type of pivot. Do not rebuild your product until you have tried selling it to a different audience. The problem is probably not the product. It is the person you are selling to.
- ▸Treat your identity as separate from your idea. The founders who fail most are the ones who cannot separate who they are from what they are building. Your startup is not you. It is a hypothesis. You can love the process without loving the plan.
- ▸Pivot early and pivot often. The window for a successful pivot is the first 12 months. After that, the odds drop to the same as never pivoting. If you have been working on the same idea for more than a year and it is not working, it is not going to start working. The data is clear. The founders who win are not the ones who refuse to change. They are the ones who change first.