Section 1
The private story sets the option set
A founder who has decided she is a builder and not a seller will not consider taking the sales lead role herself, even when the company plainly needs it. The constraint feels like a fact about her. It is a claim, formed from a small number of early experiences, that has never been re-examined. These identity claims are efficient. They save you from re-deciding everything daily. They also age badly, because they were formed under conditions that no longer hold: a different product, a smaller team, a version of you with less practice. The useful exercise is to write the claim down as a sentence and ask what evidence would change it. Doing that in front of an audience is a different skill, addressed in [How to Overcome Fear of Public Speaking Through Storytelling](/blog/how-to-overcome-fear-of-public-speaking-through-storytelling).
Section 2
Hindsight is a poor editor
Once you know the outcome, your memory rewrites the reasoning that led to it. Decisions that worked acquire a confidence they did not have. Decisions that failed acquire warning signs that nobody actually saw. The result is that experience accumulates without judgement improving much, because you are learning from an edited transcript. The fix is unglamorous: record the reasoning before the outcome exists. Not a diary of feelings. A short, dated note of what you believed, what you expected to happen, and what would tell you that you were wrong. It is the only version that hindsight cannot quietly revise.
Section 3
A decision loop worth running
Write the prediction, wait for the result, then compare the two without softening either. The gap between them is where growth actually happens. Talking through those gaps in public is a demanding format, examined in [Using Podcasts as a Medium for Business Storytelling](/blog/using-podcasts-as-a-medium-for-business-storytelling).
Section 4
Running it across a quarter
Pick the five decisions this quarter that are hard to reverse: a hire, a pricing change, a market you enter, a product line you kill, a partnership. For each, write three sentences before you act. What I believe is true. What I expect to see by a specific date. What would prove me wrong. At quarter end, read them in order. Score each on whether the prediction was right, not on whether the outcome was good, because those are different things and only one of them is a skill. Then do the identity pass. Highlight every sentence containing a claim about who you are rather than about the market. Those are the ones to interrogate next quarter.
Section 5
Traps to watch for
The redemption arc is the biggest one. Founders learn that struggle followed by triumph is a satisfying shape, and start unconsciously arranging their self-assessment to fit it. Real learning is usually messier and less resolved, and forcing the arc means discarding the parts that had the lesson in them. The second trap is publishing. The moment a private reflection becomes content, its purpose changes. You start writing the version that reads well. Keep the working record private and treat any published version as a separate artefact with a separate standard. The third is optimising the story instead of the system. Pattern-matching on your own narrative can be as misleading as pattern-matching on data, which is a problem [Using AI for Real-Time Fraud Detection](/blog/using-ai-for-real-time-fraud-detection) deals with in a different setting.