Where we start
This course runs on one talk. Daybreak, a coffee subscription company, is launching its autumn roast, and the founder gives the roastery team nine minutes on where things stand. About 560 words are said. Twelve of the things said matter. Every session captures the same nine minutes a different way, so the only thing that ever changes is what you kept and how it sits on the page.
★ What you walk out with tonight. One minute of somebody talking, captured on one page as words and marks rather than sentences, plus a two-sentence answer to "does this actually work?" that names which of three literatures you are quoting and what each one found.
Two neighbours, so you land in the right place. Giving notes a radial shape after the meeting, with one centre and keyword branches, is learn-mind-mapping-with-phoebe. Turning a finding into a deck that changes a decision is learn-visual-storytelling-with-phoebe. Here the subject is the page you make while the speaker is still talking, and the choosing that makes it.
Listening with a pen
5 min liveMike Rohde's definition is the working one: rich visual notes made from a mix of handwriting, drawings, hand-lettering, shapes and visual elements like arrows, boxes and lines. The definition lists six ingredients, and five of them are marks anyone can make. The sixth is handwriting. There is no ingredient called talent.
LiveWhat a sketchnote is, in four moves3 min▶
Take away the coloured pens and the word "creative" and the practice is four moves, each of which you already know how to make.
- Listen for the idea, not the sentence. The founder said forty words about Guji. The idea is "Guji short, Sidamo carries". Everything else was the speaker getting there.
- Give it a word. One or two, in capitals so they read across the room. The word is a handle, not a summary: it brings the forty words back when you look at it.
- Give it a mark. A box, a line, an arrow, a dot. The mark says what kind of thing it is and what it touches. Rohde builds every icon from a circle, a square, a triangle, a line and a dot, and session 2 is that alphabet. Tonight, a box is enough.
- Leave room. The speaker is not finished. A page that is full at minute three has nowhere to put minute seven, and minute seven is where the founder said nobody owns the checklist.
The first move is the whole skill. Rohde's own account of why it works is the plainest: when you try to write down every word, you stop thinking, and the limited space forces you to decide what matters. The pen is not doing anything the keyboard could not. The deciding is.
Self-studyWhere it came from, and what it is not4 min▶
The name is Mike Rohde's. In the winter of 2006 his conference notes had become, in his words, "a fanatically detailed, text-only death march." In January 2007, at a Chicago conference, he took a pocket notebook and a gel pen and decided to capture only the big ideas, drawing where a drawing was faster than a sentence. The Sketchnote Handbook (2012) turned the habit into a method, with seven chapters that run from listening to layouts to lettering. The Workbook (2014) applied it to food, travel, projects and processes.
Three neighbours share the ground and are not this. Doodling, in Sunni Brown's phrase, is "markings to help a person think", and her "strategic doodle" is the version that tracks what somebody is saying; sketchnoting is a strategic doodle with a structure. Graphic recording is the same act at wall scale, for an audience, by a professional hired to do it; it is a different job. Mind mapping gives notes a radial shape with one centre, usually after the meeting; a sketchnote can be radial and usually is not, and it is made during.
A product lead at Daybreak sat through the founder's talk with a laptop and left with 286 words, most of them the founder's. Two days later she could not find the bit about the checklist, because it was never typed; she had still been on the pricing paragraph when it went past. The person next to her had twelve boxes on one page, the checklist box was circled, and it took her four seconds to point at it. Neither of them can draw.
"But I cannot draw"
4 min liveThis is the only reason knowledge workers refuse the habit, and it rests on a misreading of what the page is for. Rohde puts a scale in the Handbook that settles it: every sketchnote sits somewhere between pure art and pure structure, and the useful ones sit toward structure.
LiveThree answers to the objection, in order of usefulness3 min▶
First: the page is not for anyone else's eyes yet. Rohde's line is that sketchnotes are about hearing and capturing meaningful ideas, not how well you draw. A box with GUJI in it does the same work as a beautifully rendered coffee bean, and takes a tenth of the time. Session 5 is about making the page readable by a colleague; tonight it only has to be readable by you.
Second: the drawing that matters is five shapes. Rohde's Handbook builds every icon from a circle, a square, a triangle, a line and a dot. Dave Gray's visual alphabet, which Sunni Brown also teaches, has twelve. Either way it is an alphabet, not a talent, and session 2 is an hour with it.
Third: time spent drawing is time not listening. This is the one nobody says. A lovely icon costs twenty seconds, and in twenty seconds the founder says the batch size. Session 3 puts a number on that cost. The art end of the scale is not wrong; it is expensive, and you are paying in ideas.
That seeded sentence is thirty-one words. "LEGAL SIGNS FIRST" is three, and with "origin claim" underneath it, everything else comes back, including the expensive letter. Most spoken sentences cut to two or three words with nothing lost, because most of what people say out loud is the run-up.
Three literatures, kept apart
6 min liveEvery book on visual notes cites "the research". It is three separate bodies of work with three different findings, and only one of them has been replicated at scale. Quoted as one thing, they promise more than any of them found. Quoted as three, they say something narrower and more useful.
LiveThe honest claim, in two sentences2 min▶
Typing makes you transcribe, and transcription is the problem. Mueller and Oppenheimer measured laptop notes at 14.6% verbatim overlap with the lecture against 8.8% for longhand, with a large effect (d = 0.94), and more overlap predicted worse conceptual answers. When Urry and 87 co-authors re-ran the study in 2021, that part replicated cleanly (12.97% vs 8.13%, g = -0.85). The part that did not replicate was the conclusion: the longhand group did no better on the quiz, and on conceptual questions the direction flipped.
So the case for sketchnoting is the behaviour, not the pen. A sketchnote forces selection because there is no room on the page for a transcript. That is the mechanism the evidence supports, and it is why session 3's bench measures verbatim overlap and ideas kept rather than pretending to measure memory. Anyone who tells you visual notes improve recall by a percentage is quoting one of the three columns as if it were the other two.
Self-studyWhat the sketchnoting studies themselves found, and did not measure4 min▶
Studies of sketchnoting as a practice, rather than of doodling or drawing single words, are few, small and mostly not about recall. The best-run one is Ohmes, Schlesier and Moschner (2024): 99 fifth-graders, five German grammar lessons, sketchnote group against a writing group, full text read. Negative emotions fell in the sketchnote group and not in the control. The paper claims no achievement or recall outcome, and that absence is the finding. Tidy, Burnham and Elkington (2022) had forensic-science students sketchnote after each lecture hour and share the pages on a board with feedback; students called it "Helpful", "Refreshing" and "Creative", and the exam analysis reports "a marked increase in the lower grade boundaries" with no control group and no effect size.
So the single most honest sentence available is this: no fetched study tests sketchnoting as taught against ordinary notes on a work-relevant outcome, with a control group and a reported effect size. What exists is mechanism evidence, and it points one way: selection beats transcription; drawing is an encoding act; a light task protects attention. The practice is built on those three, and it should be sold on those three.
Sunni Brown's 2011 article says doodling unifies three learning modalities. Her TED talk the same year says four, "with the possibility of an emotional experience." Same author, same year, same claim, one more modality in the retelling. Every secondary source on this subject inflates in exactly that way, which is why the numbers on this page come from the papers and not from the books that cite them.
One minute, one page
★ 22 min · everyone capturesA neighbour talks for sixty seconds about the last thing they worked on. You capture it with one pen on one page, and the rule for the whole minute is: no sentences, and no drawing that takes longer than a box. Then swap.
Pick the speaker. They talk for one minute, uninterrupted, about their current project: what it is, what is stuck, what happens next. Real material, not an example. You do not ask questions until the minute is up.
Capitals and boxes. Every time you hear a thing that matters, write its handle in capitals, two or three words, and draw a box around it. Leave a finger's width of white between boxes. Write the detail under the box in small letters only if it is a number or a name.
One arrow. When two of the boxes depend on each other, join them with one arrow. Only one per page tonight. The arrow is the first mark that says something a sentence could not say faster.
Read it back to the speaker from the page alone. Anything you cannot reconstruct was given the wrong handle. Anything they say you missed is the cost of whatever you were drawing at the time. Both are findings; write them at the bottom of the page.
Swap and repeat. Second time round, notice how much earlier you stop writing each idea. The speaker's third sentence is nearly always the run-up to the fourth.
One pen, dark, medium tip. Not a pencil; pencils invite fixing. Not a fine liner; you will draw too small and too much.
Turn the page sideways and put the speaker's name and the date in the top corner before they start. That corner is the only thing on the page you write before listening.
Write the handle, box it, stop. The box is drawn after the word so it fits the word. If you draw the box first you will write to fill it.
Numbers and names go under the box, small. Everything else the speaker said about that idea is what the handle is for.
Add one arrow when the dependency is clear, and only then. An arrow drawn early is a guess wearing a mark.
Count. Boxes on the page, ideas the speaker says you missed, and seconds you spent on the longest mark. Bring the three numbers to session 2, where the third one becomes the whole subject.
If you take one thing from tonight. The page is not the talk made pretty. It is the talk reduced to the things you will need, made while the talk is still happening, and the reducing is the work. A page of twelve ugly boxes that kept twelve ideas has done more than a page with one lovely drawing on it.
Before session 2
25 min- Capture one real meeting or one talk this week with the tonight rules: capitals, boxes, one arrow, no sentences. Photograph the page. Count the boxes.
- Read the page back the next morning without the meeting in your head. Mark every handle that does not bring its idea back. Under a quarter is good for a first page.
- List twenty nouns from your own work that come up in every meeting: the systems, the teams, the artefacts, the recurring problems. Session 2 turns that list into your icon vocabulary, and it needs your nouns, not coffee ones.
- Write the two-sentence honest claim for a colleague who has only read one of the three headlines. Name the three literatures, say what each found, and say which one the practice actually rests on. Under eighty words.
Three questions
Clear all three and session 1 gets stamped in your passport.
1 · A colleague says "I would sketchnote but I cannot draw." What is the accurate reply?
Rohde's definition lists six ingredients and none of them is talent. His art-to-structure scale puts working notes right of centre, where every mark is a box, a line or a capital letter. The honest addition is the cost: a rendered icon takes twenty seconds, and in twenty seconds the speaker has moved on.
2 · Someone quotes "students who write by hand learn better than students who type, Psychological Science 2014." What is the honest reply?
Mueller and Oppenheimer's mechanism variables were large and clean, and Urry et al. reproduced them in 2021 (verbatim overlap 12.97% vs 8.13%, g = -0.85). The learning outcome did not reproduce: conceptual g = -0.13, statistically equivalent to zero. Both halves, every time; the second half is what sketchnoting actually rests on.
3 · Which claim about visual notes can be defended from the fetched research, and which cannot?
Three literatures, three narrow licences. Andrade licenses a claim about attention under a dull stream; Wammes licenses one about encoding single words; the handwriting literature licenses one about transcription. None of them is a study of sketchnoting, and the two studies that are (Ohmes 2024, Tidy 2022) measured emotions and satisfaction, not recall. Quote what was found.
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 page of boxes and one arrow. Session 2 is the alphabet: Rohde's five shapes, what each one is for, and an hour building the twenty icons for your own work so that when the founder says "legal" your hand already knows the mark. Bring the twenty nouns.