How to read these
Each example below is shown as its first level only — the three-to-six branches that carry the structure — plus a note on what makes it work and where people usually get it wrong.
First level only is deliberate. Published examples are normally shown fully expanded, which is misleading twice over: it hides the decision that actually mattered, and it implies you should arrive at that much detail immediately. You should not. The first level is where a map succeeds or fails.
Copy the shape of the reasoning, not the labels.
Example 1 — Changing careers
Centre: "How do I move into work that doesn't require night availability?"
- Skills that would make me hireable elsewhere
- Evidence I can show someone
- Financial runway for a gap
- People who would refer me
- The conversation with my current employer
Why it works: each branch fails for its own reasons. You could have the skills and no evidence, or both and no runway. Compare with the obvious version — CV, job boards, interviews, networking — which is a sequence of activities, not a set of conditions, and tells you nothing about which one will sink the plan.
Where it goes wrong: people make "update CV" a first-level branch. It cannot fail independently; it fails only because the evidence branch failed. It is a Todo inside that branch.
Example 2 — Learning a language for real use
Centre: "How do I get to holding a real conversation?"
- Listening to speech at natural speed
- Producing sentences under pressure
- Vocabulary for the situations I'll actually be in
- Regular contact with a real person
- Tolerating being bad at it
Why it works: the last branch is the one no template includes and the one that most often ends the project. Naming a psychological obstacle as a branch makes it something you can attach actions to instead of something you experience as personal failure.
Where it goes wrong: organising by resource — app, textbook, tutor, podcast. That structures the map around what you bought rather than what you are trying to be able to do, and the map cannot then tell you when something is finished.
Example 3 — A side project you keep not starting
Centre: "Why has this not moved in six months?"
- What I'm actually unsure about
- The smallest version that could exist
- Time it would genuinely need
- What "done" would look like
- Who it is for, if anyone
Why it works: the centre is a diagnostic question, not a plan. When something has been stalled for months, a plan is the wrong instrument — the plan is not the missing piece, and building another one is a way of avoiding the first branch.
Where it goes wrong: mapping the project instead of the stall. You end up with a beautiful build plan for a thing you still will not start.
Example 4 — A decision with no obvious answer
Centre: "Should we move cities?"
- What would definitely improve
- What we would definitely lose
- What we're guessing about
- What is reversible and what isn't
- What has to be true for this to be a mistake
Why it works: the third and fifth branches. Separating known from guessed stops the map presenting speculation with the same confidence as fact, and the last branch is a pre-mortem — deliberately looking for the failure case, which the rest of the structure will otherwise conceal.
Where it goes wrong: pros and cons as the two first-level branches. That is a list with a line down the middle, and it produces a decision by counting rather than by weight.
Example 5 — Studying for an exam
Centre: "What do I need to be able to do on the day?"
- Topics I can already do under time pressure
- Topics I understand but can't do quickly
- Topics I have not actually attempted
- Exam mechanics — format, timing, marking
- Recovery from a bad practice run
Why it works: split by state of readiness rather than by syllabus. The syllabus is the textbook's structure; this is yours, and it directs effort at the second and third branches where effort actually changes the outcome.
Where it goes wrong: copying the syllabus contents as branches. It feels thorough and spreads attention evenly across material you already know.
The pattern across all five
Every good example splits by where the uncertainty lives, not by topic. Topic splits are easy to generate, look organised, and tell you nothing about what will go wrong. Uncertainty splits are uncomfortable to write and immediately useful.
The other shared trait: each has a branch most people would leave out — tolerating being bad at it, what has to be true for this to be a mistake, recovery from a bad practice run. Those are the branches that stop a map being a tidy description of the easy parts.
Building your own
Take whichever example is closest, throw away the labels, keep the reasoning. Write your centre as a question, then ask what could fail on its own terms.
You can build one now without an account — the real workspace in your browser — or start from a template if a blank canvas is the obstacle.