By the time your company has been closed for a month, you have told the story many times. To a co-founder. To an investor. To a peer who reached out. To family. The story has found its shape: the sequence, the turning point, the call you made or did not make. The rough edges have been sanded. It flows.
That smoothness is not evidence of accuracy. It is evidence of repetition. And the mechanism that converts repetition into felt truth is not one you have much say over once it is running.
Why the version you keep telling feels correct
Kahneman's research on how thinking operates draws a line between two modes: the fast, automatic system that runs most of your day without effort, and the slow, deliberate system that reasons carefully but tires quickly. The fast system reads a quality called fluency: how easily something comes to mind. It uses fluency as a proxy for truth. A story you have told many times runs without friction. That smoothness is interpreted, automatically and below the level of your awareness, as accuracy. The polished version of what happened does not just feel practiced. It feels true.
This is not a character flaw. It is how the machinery works. Repeated rehearsal does not make a story more accurate. It makes it more fluent. And fluency reads as truth to the part of your mind that is doing most of the driving.
What the story is actually carrying
Seligman's research on explanatory style identifies three dimensions in how people explain bad outcomes. The first is permanence: whether the cause is seen as temporary or fixed. The second is pervasiveness: whether it is specific to this situation or universal, spreading across every domain. The third is personalization: whether the cause is located inside you or in the situation around you.
A pessimistic explanatory style treats bad events as permanent, pervasive, and highly personal. A more accurate reading of most shutdowns is mixed: some causes were internal, some structural; some were fixable, some were not; the failure was real but not a universal verdict. The honest post-mortem of most companies lands somewhere in the middle of all three dimensions.
The problem is that the story you are rehearsing was formed when your reasoning was most compromised. The narrative crystallised in week one or two, under exhaustion, under the immediate press of the close, when your slow deliberate system had the least in the tank. That version, right or wrong on the dimensions that matter, is the one that is now fluent. Every subsequent rehearsal is not reviewing it. It is deepening it.
Habitual retrieval, not deliberate recall
Wood's research on habit and cue-response patterns shows that behaviours repeated in consistent contexts become automatic: the cue triggers the response without a review step in between. The social situation of being asked "what happened with the company?" is a cue. After enough repetitions, the answer is not recalled and examined. It is retrieved. The response runs before you have chosen it.
This is worth pausing on. You may believe you are still examining the story each time you tell it, catching it up against new evidence, adjusting it as your perspective shifts. In the early weeks, maybe. After fifteen repetitions, probably not. The version that arrives when the question is asked is the habitual one, and habitual responses do not pause for review.
The decisions you will build on top of it
The narrative is not just a story about the past. It is the frame through which you will read the next opportunity, the next team dynamic, the next strategic call.
If the story carries a permanence claim ("I am not someone who can close enterprise deals"), that attribution shapes your search radius for the next move. You will avoid what you attributed the failure to, whether or not that attribution is accurate. If it carries a pervasiveness claim ("I make bad calls under pressure"), that belief will sit behind your confidence in a whole category of future decisions, even ones where the shutdown tells you nothing relevant.
Sheeran and Webb's research on the gap between intention and action shows that intentions formed from a depleted or distorted state have lower follow-through. Plans built on an inaccurate self-story are more likely to collapse: not because of willpower, but because the premise was wrong from the start. You can form the best implementation plan in the world; if it is grounded in a story that overstates your permanent limitations or underweights the structural causes, the plan is working from a corrupted input.
Three questions before the next rehearsal
The goal is not to revise the story into something more flattering. It is to check whether what you are cementing is worth cementing.
First: does the story distinguish between causes you controlled and causes you did not? Complex failures mix both categories, and a story that loads everything onto one of them is almost certainly compressed. Edmondson's research on failure taxonomy identifies complex failures as those with multiple contributing causes, most of them not individually preventable. Most shutdowns qualify.
Second: does the story treat any cause as universal? "I always do this" or "this always happens to me" are permanence and pervasiveness claims that need evidence, not just a feeling formed in the worst week.
Third: is the level of personalization accurate? Some of the cause was yours. Some was structural, timing-dependent, or distributed across a system too complex for any one person to have caught. A story that locates all of it inside you is almost certainly wrong, and Seligman's research says that running it is costly: it predicts passivity, avoidance, and a narrowed search radius going forward.
You do not need to dismantle the story. You need to look at it once, clearly, before the next dozen rehearsals push it past the point where it can be examined.