The Founder's Decision Log Is Worth Ten Minutes

A hedge fund manager asked Daniel Kahneman what one thing he could do to make better decisions. Kahneman's answer was to walk to a drugstore and buy a cheap notebook.

Write down what you expect to happen, why you expect it to happen, and how you feel while you're deciding. Do the thing. Read the entry later.

The most decorated psychologist in the field, asked for the highest-leverage habit in decision making, pointed at stationery.

Memory rewrites itself. Once you know how something turned out, your brain edits the story of how you decided: the doubts get retrofitted, the hunches get promoted to convictions. Kahneman's own phrasing: we look back on our decision-making process and tilt it in a way that looks more favorable to us.

Founders pay for that twice. First when a call goes wrong and the lesson is unrecoverable, because you can't tell whether the reasoning was bad or the luck was. Second when the same call comes back around.

Pricing. A hire. The customer who pays late and complains early. Six months on, the option you rejected has lost its reasoning, so you re-run the whole debate and land in the same place you already landed.

Shane Parrish's decision journal template is the cheap fix. Before you act, write down the situation and how you're framing it, the variables that matter, the alternatives you seriously considered and why you rejected each one, the range of outcomes, and the outcome you actually expect with a probability attached.

Then add how you feel physically and mentally, because a call made at 11pm on four hours of sleep should carry that label.

Two entries on that list do the real work. The alternatives you rejected, because the tradeoff is what a future reader needs. And the probability, because "I had a feeling" can never be checked and "70%" can.

Writing also changes the decision before it's made. You sit down to record your reasoning on a hire and find the reasoning is two sentences long and one of them is "we need someone." The entry turns into the finding.

When the call belongs to a team, Gary Klein's premortem covers the same ground from the other direction. Assume the project already failed. Explain why. Klein's observation in Harvard Business Review: projects fail at a spectacular rate partly because people who can already see the crack stay quiet during planning. Imagining the failure gives them a safe way to say the thing they came in already believing.

Engineering teams built this into their tools in 2011. Michael Nygard published the architecture decision record: a short numbered file kept beside the code, one record per decision, listing the forces in play and the choice made. The why lives with the how.

His argument is the one a founder needs. Without the motivation on record, anyone who arrives later can only accept the decision blindly or change it blindly. A project that accumulates enough of that stops being able to move. Superseded records stay in the file, marked as superseded. The archive is the asset.

Ray Dalio pushed it further and made it the operating system of a company whose culture he describes as an idea meritocracy built on radical transparency. The written reasoning is what gets challenged. Deferring to whoever holds the title is the failure mode.

The compounding is what makes the ten minutes worth it. Forty entries in, you stop learning about individual decisions and start learning about yourself: that you lowball hiring ramp by two months, that your urgency decisions underperform your patient ones, that you say 80% when you mean 60%.

Nobody sells you that feedback, and memory never produces it.

The earliest entries are also the most valuable, because the calls you make when the company is small set the defaults that last for years, and that reasoning exists nowhere else.

Ten minutes before the call, or the same argument again in six months with less information and more confidence. That's the trade, and it isn't close.