Founder Psychology6 min read·

    Why Founders Mistake Enthusiasm for Traction

    Enthusiasm from early customers is the most dangerous feeling in startups. Not because it is wrong — but because it is easy to confuse with traction, and the founder's brain is built to make that confusion as comfortable as possible. Every startup has a phase where the feedback feels positive, the demos land, the early conversations are electric. The question is whether that phase corresponds to something real or whether it is a temporary state produced by selection bias and social dynamics.

    The enthusiasm trap

    Enthusiasm is cheap to generate in a sales conversation. A skilled founder who believes in their product — and most founders do — naturally runs conversations that produce enthusiasm. They lead with the most compelling framing. They respond to objections in ways that feel satisfying. They talk to people who already have the problem the product solves. The result is a string of conversations where everyone seems excited, and the founder walks away believing the market wants what they built.

    The problem is that enthusiasm in a 45-minute conversation does not predict purchase, retention, or recommendation. It predicts that the person enjoyed the conversation. You can have twenty enthusiastic conversations and not one sale, or twenty sales that churn within 90 days. Enthusiasm is a social signal, not a market signal.

    Traction is the opposite: cold, structural, hard to fake. It shows up in renewal rates, word-of-mouth referral volumes, expansion revenue, and the inbound conversations you didn't initiate. When traction is real, it is almost boring. The numbers do not require the founder to interpret them generously.

    Why the founder's brain amplifies it

    Confirmation bias — the tendency to weigh evidence that confirms existing beliefs more heavily than evidence that contradicts them — is the primary mechanism. Once a founder has decided that the product solves a real problem, every enthusiastic conversation becomes evidence for that belief. Every lukewarm conversation becomes noise, attributed to the wrong prospect or the wrong framing. Every non-conversion becomes a process problem, not a product problem.

    Optimism bias compounds this: founders systematically overestimate positive outcomes and underestimate how often things will go wrong. Combined with the social dynamics of early customer conversations (people don't like to disappoint a founder who clearly cares deeply about their work), the result is a feedback loop that consistently produces more positive signal than the reality warrants.

    In 74% of failed startups, premature scaling was a factor — and most premature scaling decisions are preceded by a period of enthusiastic early feedback that was mistaken for evidence of PMF.

    The replacement test

    One diagnostic that cuts through enthusiasm: ask your three most enthusiastic customers what they would use if your product disappeared tomorrow. If they can name a specific alternative immediately, your product is a preference, not a dependency. If they pause and say something like 'I don't know, I'd have to figure something out,' or 'We'd have to rebuild part of our process,' you are closer to a real need.

    The replacement test doesn't require courage in the way that asking 'would you be very disappointed if you could no longer use this product?' does. People answer the hypothetical displacement question more honestly because it's not framed as feedback — it's framed as contingency planning. You get behavioural intent rather than social approval.

    Run it in your next five customer conversations. Write the answers down verbatim. The pattern across those five conversations will tell you more about your PMF trajectory than six months of enthusiastic demos.

    cognitive biasconfirmation biascustomer feedbacktraction

    Score your own PMF in 20 minutes.

    Free PMF score across market, founder, and execution readiness — with named blind spots and specific first actions. No credit card required.