Who this is for
A decision with real consequences that you have been carrying for weeks — a move, a job, a relationship, a large purchase, ending something. You have thought about it constantly and made no progress, which is the characteristic signature of a decision being circled rather than examined.
This template exists because the default tool for this is a pros and cons list, and a pros and cons list is a poor instrument. It produces a decision by counting, when the whole difficulty is that the items have wildly different weights and wildly different certainties. Six small pros do not outweigh one large con, and a list cannot show you that.
The branches
What is actually being decided. Often the stated decision is not the real one. "Should I take this job?" is sometimes "should I leave this city?" or "am I willing to disappoint someone?" This branch is short and it frequently changes everything downstream.
What I know versus what I am guessing. Split explicitly into two sub-branches. Most decision paralysis is caused by treating speculation with the same confidence as fact — "I would hate the commute" is a guess; "the commute is 70 minutes" is known. Once separated, the guesses become a research list rather than a source of dread.
What is reversible and what is not. A genuinely underused distinction. Reversible decisions deserve speed, because the cost of being wrong is small and the cost of delay is real. Irreversible ones deserve the slow treatment. Most decisions are more reversible than they feel, and identifying which parts can be undone often dissolves the paralysis.
What I would lose that I am not counting. Every choice forecloses something, and the foreclosed things are systematically undercounted because they are currently free and invisible. Proximity to someone. A routine. An identity. Optionality itself.
What has to be true for this to be a mistake. A deliberate pre-mortem. Assume it went badly in two years — what happened? This branch surfaces the assumption the rest of the structure is resting on, and it is usually one specific belief you have not examined.
Who has actually done this. Not opinions from people imagining it — accounts from people who made the same call. What surprised them? What do they wish they had known? One real conversation here outperforms a month of internal deliberation.
What would tell me sooner. Is there a cheap test? A trial, a visit, a conversation, a month of doing it partially? Decisions that feel binary often have a probe available, and the probe is nearly always worth more than further thinking.
How to use it
Import it and mark "what I know versus what I am guessing" as Focus. Sorting that one branch resolves a surprising number of decisions outright, because much of the weight turns out to sit on guesses that are cheap to check.
Use relationship lines rather than hierarchy for the dependencies — the pre-mortem branch usually links to a specific guess, and that link is the most useful line on the map.
Set a Deadline for the decision itself if a real one exists, and if it does not, set an Aim date anyway. This is the exception to the usual advice about invented dates: a decision without a horizon can be circled indefinitely, and the circling is itself costly.
What to add once it is yours
Put the real specifics in — the actual numbers, the actual names, the actual thing you are afraid of. This template is close to useless in the abstract.
Attach Proof to the research: the conversation you had, the number you checked, the trial you ran. Watching guesses convert into knowns is what makes the map feel like progress during a period that otherwise feels like paralysis.
Signs it is working
- The guessing branch is shrinking as things get checked
- You have talked to someone who actually did it
- You know which parts are reversible
- The real decision turned out to be slightly different from the stated one
Signs it is not
- The map is a restructured pros and cons list
- Nothing has moved from guessing to known
- You are researching to postpone rather than to decide
- The pre-mortem branch is empty