Where we start
This course runs on one meeting. Daybreak, a coffee subscription company, is planning its autumn roast launch, and fourteen things were said in forty minutes. Every session takes those same fourteen things and gives them a different shape, so the only thing that ever changes is what you did with them.
★ What you walk out with tonight. One page of your own notes redrawn as a map with a centre, three to seven branches and one word per branch, plus a two-sentence answer to "does this actually work?" that you can defend with numbers.
Two neighbours, so you land in the right place. Drawing icons and lettering while somebody talks is learn-sketchnoting-with-phoebe. Writing a map as Mermaid code is learn-mermaid-with-phoebe session 5. Here the subject is the shape itself: what goes in the middle, what becomes a branch, and how you decide.
A list is the order things were said
5 min liveNotes taken live are a transcript with the boring bits left out. That is fine for capture and useless for everything after, because the order of a conversation is the order two people happened to think of things, and it will never be the order you need to find one of them again.
LiveWhat a mind map is, in four parts3 min▶
Strip away the coloured pens and the word "creativity" and a mind map is a very specific object. Four things make it one, and dropping any of them makes it something else.
- One centre. A single idea in the middle of a landscape page. Not a title, an idea: "Autumn launch", not "Notes from Tuesday".
- Branches that radiate. The main themes leave the centre as thick lines, and everything else hangs off them. Position means membership: if it is on the Supply branch, it is a supply matter.
- One keyword per line. Not a sentence. The word that brings the sentence back. "Guji short" does the work of "Guji supply is still short so the autumn blend leans on Sidamo", and it does it in a glance.
- Hierarchy by distance. Big ideas near the centre, detail further out. You can read a good map at three depths: the centre alone, the five branches, or everything.
The first and third are where most maps fail. People put a title in the middle because that is what goes at the top of a page, and they write sentences on the branches because that is what notes are. Both are the list creeping back in.
Self-studyWhere it came from, and what to do with the folklore4 min▶
The radial diagram is old; Porphyry's tree of categories is third century. The modern packaging is Tony Buzan's. He popularised the term through a 1974 BBC series, Use Your Head, and its companion book, then codified the method in The Mind Map Book (1993) and restated it in Mind Map Mastery. The laws in part 2 come from there.
Two things travelled with the method that are worth separating from it. The first is the claim that a map "uses both hemispheres" and mirrors how the brain stores ideas. That model of the brain has not held up, and the method does not need it: the laws work because they are layout rules, and layout rules work on paper regardless of what neurons are doing. The second is a trademark. "Mind Map" was registered by Buzan's organisation; the diagram itself is nobody's property and the generic term is in universal use.
A product manager at Daybreak keeps a one-page map for each supplier. When the Guji shortage hit, she did not search her notes; she looked at the Guji page, followed the Supply branch to "backup: Sidamo", and had the answer in the time it took to focus her eyes. The map had been drawn eight months earlier by someone who no longer worked there. That is the whole value proposition, and it needs no neuroscience.
The ten laws, read as craft rules
5 min liveBuzan's books give ten laws. They are usually presented as things the brain likes. Read them instead as what a typographer or a map-maker would tell you, and they become easier to follow and easier to defend, because each one solves a visible problem on the page.
LiveThe three that matter most, and the one people resist3 min▶
If you keep only three: one keyword per line, one colour per branch, white space between groups. Those three turn a page you have to read into a page you can look at. The rest are refinements.
The one people resist is the keyword rule, and the resistance is honest: a keyword feels like losing information. It is not. The keyword is a retrieval cue, and the sentence it stands for comes back when you read it, because you wrote it. What you cannot do with a sentence is hang three more things off it, and that is the operation a map exists to perform.
The one people over-apply is number ten. Emphasis is relative, so a map where everything has an arrow has nothing emphasised. One box, one arrow, one oversized word per map is usually the budget.
That seeded line is twelve words. "Guji short" is two, and it brings back the other ten plus the Sidamo detail, which becomes its own twig. Most captured lines cut to two or three words with nothing lost, because most of the words in a note are grammar holding the facts apart.
What the research says, both halves of it
5 min liveMind mapping has a large literature and a lot of it is positive. It also has a habit of being quoted by its headline number only. Read the whole abstract and a more useful picture appears: a real effect, a narrow honest claim, and two findings the enthusiasts leave out.
LiveThe honest claim, in two sentences2 min▶
Making a map is an active, organising thing to do with material, and active organising reliably beats passive re-reading. Whether the radial shape adds anything on top of the organising is not established: the best-designed comparison found maps only slightly ahead of an outline of the same text.
That is a smaller claim than the books make and a much stronger one than "it works for me", because you can hold it against a sceptic. The people who dislike maps are usually reacting to the neuroscience wrapper. Drop the wrapper and the objection usually goes with it.
Self-studyWhy the numbers are so large and so shaky at once4 min▶
A standardised difference of 1.45 is enormous; in education research an effect of 0.4 is considered worth having. Numbers that big should make you suspicious rather than excited, and the 2026 meta-analysis says why in its own limitations section.
- Nobody was blinded. In 98% of the trials the students knew which group they were in and so did the examiners. Enthusiasm for a new method shows up in scores.
- The control was a lecture. Almost any active method beats sitting still. The interesting comparison is against another active method, which is exactly the comparison where the advantage shrinks to "slight".
- Heterogeneity above 80%. The studies disagree with each other more than they agree, which usually means they are not measuring one thing.
- Publication bias. When the authors corrected for the studies that probably went unpublished, two of the secondary outcomes stopped being significant.
The 2024 nursing study is the one to remember, and not for the reason its authors intended. It ran two topics across 112 students with the groups swapped, and its abstract reports mind mapping superior on both. Its own Table 1 shows the mind-map column ahead on the first topic's gain (5.42 against 3.55) and behind on everything else: the second topic's post-test, its gain (3.57 against 4.57), and one-month retention on both topics. Either a column label is wrong or the abstract is reading the table generously; the paper does not say which. The students' Likert responses showed "widespread agreement" that mapping helped. Feeling helped, being helped, and being reported as helped are three different measurements, and the table is the only one of the three you can check.
Redraw one page of your own notes
★ 22 min · everyone drawsReal notes from a real meeting this week, not an example. Paper turned sideways, three pens, and the rule for the next twenty minutes: no sentences on the page.
Find the centre. Not the meeting's title. The thing the meeting was for. If your notes were about three things, you have three maps, and you draw the one that matters this week.
Name the branches. Read the whole page once and write three to seven headings on scrap paper before you draw a line. If you have nine, two of them belong together. If you have two, one of them is hiding three.
One word each. Draw the branches, thick, one colour each, and print the heading along the line. Then go line by line through the notes and cut each one to the word that brings it back.
Hang the detail. Numbers, dates and names go on thinner lines off the keyword they belong to. "4000 bags" hangs off "First batch", not off the centre.
Read it back to the person next to you, from the map alone, without the notes. Anything you cannot reconstruct was cut to the wrong word. Fix the word, not the rule.
Turn the page sideways before you write anything. A portrait page has room for two branches and a list; a landscape page has room for six.
Write the centre in a box, in capitals, in the middle. Two or three words. If you can add a small picture, do; if you cannot draw, a box in a second colour does the same job.
Draw the main branches thick and curved, one colour each, spaced around the centre with gaps between them. Print the branch name along the line so the line is as long as the word.
Move every note onto a branch as a keyword. Anything that does not fit a branch either needs a new branch or was not worth writing down. Both are findings.
Add one piece of emphasis, and only one. Box the thing that needs a decision, or arrow the two twigs that depend on each other. Then stop.
Count. Branches off the centre, words on the longest label, and the number of notes that would not fit anywhere. Bring the three numbers to session 2, where they become the whole subject.
If you take one thing from tonight. The map is not the notes made pretty. It is the notes moved to where they will be looked for, and the moving is the work. A map drawn in five minutes that puts every fact under the right branch has done more than an hour of colouring.
Before session 2
25 min- Map one more meeting, live this time. Draw the centre and guess the branches in the first two minutes, then capture straight onto the map. Note where the guess was wrong; that is the real lesson and it is different for everyone.
- Redraw tonight's map from memory tomorrow morning, without looking. Compare. The branches you forgot were badly named.
- Find one confident claim about mind mapping online and check whether it quotes both halves of its source. Bring the claim and the missing half.
- Count your longest label on both maps. If it is over three words, cut it and see whether you can still read the map back a day later.
Three questions
Clear all three and session 1 gets stamped in your passport.
1 · A colleague writes "Guji supply is still short so the blend leans on Sidamo" along a branch. What is the problem?
The keyword rule is not about tidiness. A sentence is a leaf: nothing can hang off it, so the next fact about Sidamo goes on a new line and the map becomes a radial list. "Guji short" with a twig for "Sidamo blend" carries the same facts and can still grow.
2 · Someone quotes the 2026 meta-analysis: "52 trials, scores up 1.45 standard deviations". What is the honest reply?
The number is real and the paper says so. The same paper says 98% of its trials were unblinded, the control was passive lecturing, heterogeneity was above 80%, and certainty was graded very low. Nesbit and Adesope's earlier synthesis, which compared maps against other active methods, found the advantage shrinks to slight. Both halves, every time.
3 · Why do the ten laws hold up even though the brain model that came with them does not?
The laws are typography and cartography, not neuroscience. A map-maker would give you most of them unprompted. That is why they survive the loss of the hemispheres story, and it is also the version you can defend to a sceptical colleague without an argument about the brain.
What this session teaches, and where it came from
Every fact on this page, and its verification tier, is recorded in the course's source map.
Session 1 cheat sheet · pin this
You have a map with some number of branches and some longest label. Session 2 is about both numbers: why four and seven are both quoted as the limit, what the argument between them is actually about, and the keyword rule taken seriously enough to run on real notes.