A mind map is a page with the topic in the middle and branches radiating out, sub-branches off those, one idea per branch. Tony Buzan popularised the format in the 1970s and the basic shape has barely changed since, because it is a good fit for a real problem: some material is a web of relationships, and a list forces it into a line.
The technique is oversold and underused at the same time. Oversold as a general note-taking method, where it is often the wrong tool. Underused for the two jobs it is genuinely excellent at.
How to make one
1. Topic in the centre, on a page turned sideways. Landscape gives branches room to grow right and left. Two or three words, not a sentence.
2. Main branches first, and stop at about five. These are the big divisions of the topic. Doing this before any detail forces you to decide the structure, which is the part that produces understanding. If you cannot name the main branches, you do not yet know the topic well enough to map it — useful information in itself.
3. One word or short phrase per branch. Not sentences. The compression is doing work: choosing the single word that captures a branch is a judgement, and judgement is what you are trying to practise.
4. Branch outwards into detail. Sub-branches attach to their parent, physically, so the hierarchy is visible without reading. Keep going until the detail is exhausted or the page is full.
5. Draw the cross-links, and label them. Lines between branches that relate. This is the step people skip and it is where most of the value is — an unlabelled line means almost nothing later. "Causes", "contradicts", "is a special case of", "only if" — a labelled relationship is a claim you can be right or wrong about, and an unlabelled one is decoration. Concept mapping's insistence on this is the single most useful borrowing available; see concept mapping.
Colour and images are optional. Use a small amount deliberately — the benefit comes from contrast, and every extra colour spends it. See color-coding notes.
Use one for these two jobs
Generating. A blank centre and empty branches ask "what are the parts of this?", which is a much better prompt than a blank linear page asking you to start writing in order. Excellent for planning an essay, scoping a project, or working out what you actually think.
Reviewing. Rebuilding a map from memory, after the fact, is a strong revision technique — one page, seen at a glance, produced by retrieval. See active recall.
Don't use one for these
Live capture in a fast lecture or meeting. You cannot know the structure while it is still arriving, and you will spend the hour deciding where things go instead of listening. Record linearly and map afterwards — see linear vs nonlinear note-taking.
Sequential material. Procedures, timelines, narratives. Anything where order is the content is worse as a map.
Anything you need to search in three years. A photographed map is not findable text. Important reference material should end up written out.
Large volumes. One page is powerful. Forty maps are unnavigable, because there is no index and no order.
The honest evidence
This is where most articles overclaim, so worth being straight about.
The best-known study is Farrand and colleagues in Medical Education (2002), which found medical students recalled meaningfully more factual material after mind mapping than after their preferred study method. Two things about it are usually left out: the advantage was modest, and mind mapping reduced students' motivation compared with what they normally did — a real problem for a technique that requires sustained use.
The wider literature is mixed. Mind mapping generally beats passive rereading, which is a low bar that almost every active technique clears. It does not clearly beat other active techniques such as retrieval practice or self-explanation. And a lot of the reported benefit is plausibly explained by the fact that making a map forces you to reorganise material rather than copy it — which is levels of processing doing the work, not the radial layout.
The practical reading: mind mapping is a good way to force elaboration. It is not magic, and if you find it tedious, another active method will serve you about as well.
The mistake that makes maps useless
A map of terms connected by unlabelled lines encodes almost nothing. It looks organised, it took an hour, and a month later it is a word cloud you cannot reconstruct the argument from.
The test: cover the map and try to state three relationships it shows. If you can only produce "these things are related to the topic", the map recorded vocabulary rather than understanding. Labelled links fix this, which is why the fifth step above is not optional.
The second failure is prettiness. Time spent on lettering and colour schemes feels like studying and is not — the illusion of competence in visual form.
Digital or paper
Paper is faster to start, has no tool decisions, and is better for thinking. Digital is better when the map will change a lot, needs to be shared, or has to be searched later — restructuring on paper means redrawing.
If you want the map to be part of a permanent archive rather than a one-off thinking aid, digital wins by default, because paper maps do not get filed and do not get found. See the best notes app for mind mapping.
The minimum viable version
Topic in the middle, five branches, detail underneath, then three labelled lines between branches. Twenty minutes.
Then close it and redraw it from memory a week later. The second version is where the learning happens, and it is the step almost nobody does.
More in mapping method, mind map vs outline, and note-taking methods.