# The Four Ways Founders Fool Themselves - Author: Petri Lehmuskoski - Published: 2026-10-09 - Group: Foundations - Book chapter: 3 - Canonical: https://morescarsthantrophies.com/writing/the-four-ways-founders-fool-themselves Three to five customers. That's what the founders had to show after chasing a dream of fourteen million potential customers worldwide. Three to five felt like proof the whole idea was too small to survive a serious conversation. I was an early backer. The company was running out of money, and that kind of pressure does something specific: it makes founders finally willing to hear what they'd been talking past for a year. An advisor outside the founding team had been pointing at a narrow group of companies, structurally different from the fourteen-million-customer dream, that looked like real buyers. Nobody had taken it seriously enough to test it. Running low on runway is what finally made them listen. Once they did, the founders did the unglamorous work: name exactly why that narrow group would buy, Buyer Proof, and what the product was actually worth to them, Value Proof. Then they pitched to that group on purpose. Seventeen companies this time, not the fantasy market. Thirteen closed. A close rate near 80 percent, once the pitch matched who was actually buying instead of who the founders had wished was buying. Once Buyer Proof and Value Proof were finally clear for that segment, Scaling Proof followed almost on its own. It took the better part of a year, most of the company's remaining runway, and an outside voice willing to say it plainly, before anyone tested it properly. Nobody was incompetent. Nobody was lying. The company simply couldn't see a segment of seventeen while staring at fourteen million. Across 10,000+ startups reviewed as an investor, I've watched this exact failure mode again and again, and it never looks like stupidity. The founders in that story were sharp. What got them wasn't a lack of intelligence. It was a founder's specific talent for turning incomplete evidence into a story that lets the company keep moving. Signals bend under pressure, and founders are not neutral observers of their own company. The old line applies here better than anywhere: the fish grows with every telling, weigh it at the dock. Trying to make something work creates its own pressure to read every report as proof that it's working. Small evidence at an early stage is not weak evidence. It's evidence that hasn't yet been amplified by scale, which is exactly the size a genuine early signal is supposed to be. Founders often wait for a version of proof that only exists once the company is already large. By the time they'd accept it, they've already spent the year they could have used to test the smaller version. That pressure runs through four specific channels. ## 1. Selective interpretation The easy version of this is discounting bad news. The harder version, the one in the story above, is discounting good news because it doesn't match the scale of the story you've already committed to. A signal that contradicts the current direction gets scrutinized. A signal that supports it gets waved through. The most reliable data usually comes from people who say no. Refusal removes politeness. It exposes what didn't move, and why. If five people turn you down for the same reason, that's not five isolated objections, it's a pattern, and it's the first honest version of the truth you'll get for free. Founders who interpret selectively aren't lying to themselves about individual facts. They're deciding, case by case, which facts are allowed to count. A rejection gets explained away. A sale too small to fit the story gets ignored. Both decisions feel reasonable in the moment, and both get defended with the same word: judgment. Neither one is judgment. Both are selection. ## 2. Momentum becomes truth Once something starts moving, it gets harder to question. Activity is visible, so it reads as progress: meetings held, features shipped, a pipeline that looks busy. Stopping to ask whether it's actually working feels like failure. Continuing feels like progress, even when nothing fundamental underneath has changed. This is what let the fourteen-million-customer story survive a year past its expiration date. The company was doing things: hiring, pitching, building. All of it supplied its own kind of proof. It felt like forward motion. It was motion around a stalled decision. The tell is simple, if uncomfortable. If you can only name what got built or shipped last quarter, and not what changed about the company's central assumption, momentum has replaced evidence as your reason for continuing. ## 3. Effort creates commitment The more time a founder has already spent on a direction, the harder it becomes to question it. The question quietly shifts from is this true to how do we make this work, and that shift moves the company from testing reality to protecting the effort already spent. This is sunk cost, but it rarely shows up as an abstract bias. It shows up as loyalty to a plan, dressed as conviction. The founders in the story weren't defending a spreadsheet. They were defending eighteen months of belief and several fundraising conversations built on the fourteen-million number. Reversing course meant admitting all of it had been aimed at the wrong target. Conviction is supposed to survive contact with evidence. When it only survives by ignoring evidence, it isn't conviction, it's a sunk cost with better branding. ## 4. External pressure accelerates decisions Investors aren't the only source of pressure. Partners want certainty, early employees want direction, and friends and family want to hear that it's working. Over time the company's actual position stays uncertain while everyone around the founder needs it to sound concrete. That gap is where burnout starts, and it's also where founders stop reporting reality and start reporting a version of it that keeps people calm. The pressure doesn't come from any one person lying. It comes from a system where saying "I don't know yet" out loud, repeatedly, to people who are counting on you, is uncomfortable enough that rounding up feels kinder, not because the founder is dishonest, but because the alternative is telling everyone who depends on the company that you don't actually know yet. None of this happens in isolation, either. By the time a weak signal reaches a founder, it has usually passed through a few people first: an account manager, a customer success lead, a co-founder trying to keep morale up. Every hop is a chance to soften it. Bad news arrives as an interpretation, the pilot's going fine, rather than a behavior, the buyer hasn't touched budget. The team isn't lying. They're responding to what gets rewarded. If the last piece of bad news got a hard reaction, the next one arrives later, and softer, or not at all. ## Back to the thirteen customers All four of these were in the room at once. Selective interpretation kept three to five customers filed as failure instead of as a first real signal, because it didn't match the size of the story. Momentum kept the company pitching and hiring toward the fourteen-million dream long after the actual traction said otherwise. Effort creates commitment explained why an outside voice naming a smaller, real segment got ignored for as long as it did: nobody wanted to admit a year of fundraising conversations had been aimed at the wrong market. And the same kind of pressure that had kept the fantasy alive, the need for a story big enough to satisfy investors and the founders' own hopes, only broke once a harder pressure arrived: the money running out. Scarcity did what analysis hadn't. It forced a straight look. None of that is stupidity. It's what happens to reasonably smart people holding evidence that's inconvenient. The fix isn't more intelligence. It's a different question, asked earlier. You don't need an angel investment story to recognize this. It shows up in the metrics review where the number everyone tracks quietly stops meaning what it used to. It shows up in the customer call that gets summarized as "went fine" instead of quoted directly. It shows up whenever a founder needs the current plan to be right badly enough that the evidence gets read to match. ## The question that catches it early There's one question that would have shortened that company's detour by most of a year: what would prove this wrong, and has that already happened? "Is it working" invites the founder's own hope into the answer. "What do the numbers say" invites whatever reading the founder needs, since numbers can be read the size you require, which is exactly what happened with three to five customers. The useful version names the specific behavior that would mean the current direction is wrong, then either checks whether it has already occurred or tests for it directly instead of waiting for it to show up on its own. In the story above, the test that would have settled it was available a year earlier: pitch the narrow segment on purpose and count the closes. Nobody ran it, because nobody had decided in advance what result would count as proof either way. When they finally ran it, forced by the calendar rather than curiosity, the answer took weeks and wasn't ambiguous at all. Write the falsifying condition down before you need it. Once you're inside the decision, you'll read every signal in whatever direction keeps you moving. Do it for whatever number you're currently explaining away. If you can't say what would prove your current read wrong, you don't have a plan, you have a preference with a spreadsheet attached. ## Founders don't die from lack of vision They die from unresolved reality: evidence that was available, sitting quietly, while the story kept moving. The fourteen-million-customer company survived its own delusion because someone from outside finally forced a straight look, and the underlying signal was strong enough to work once it got one. Most companies don't get that kind of margin for error, or that kind of outside voice, in time. That's the only version of this discipline worth building. Not a founder who never misreads a signal. Nobody does that for long. A founder who gets caught less often, and catches it sooner when they don't. The job was never to be right on the first read. It's to become harder to fool on the second one. More scars than trophies.