Robert Greene writes books by filling boxes with index cards. Ryan Holiday, who worked as his research assistant, uses a version of the same system and has said that nearly every dollar he has earned as an adult was earned first on a notecard. The method is old, analog, and unreasonably effective for one specific job: turning a pile of reading into a piece of writing.
The mechanics
It is almost too simple to write down.
- Read first, mark lightly. Fold a corner, underline, dog-ear. Do not take notes yet.
- Wait. A week or two, sometimes longer. Then go back through the book and transfer only what still seems worth transferring.
- One idea per card. A 4×6 card gets a quote, an anecdote, a fact, or a thought of your own, plus the source.
- Theme in the top corner. A word or two: stoicism, pricing, chapter 3. This is the only piece of metadata.
- File it in a box, behind a divider. When a project starts, you pull the relevant dividers and spread the cards on a table.
That last step is the whole point. The cards get sorted, resorted, grouped, and thrown out until an order appears, and that order is the outline. Greene has described going through several dozen books and ending up with hundreds of cards from which the chapters emerge.
What the card does that a document doesn't
It enforces one idea per note by physics. You cannot write four thoughts on a card without running out of room, so you don't. Digital notes have no such wall, which is why so many are three unrelated things in a trench coat. This is atomic notes achieved through stationery rather than discipline.
It delays the note until the idea has survived. The wait between reading and carding is a filter — most of what feels brilliant on page 40 does not survive two weeks. Taking notes as you read captures your enthusiasm; taking them later captures your judgement.
It makes rearranging cheap. Moving a card takes half a second and no decision. Reordering sections of a document takes a scroll, a cut, a paste, and a moment of second-guessing. When arrangement is nearly free you do it far more, and arrangement is where the thinking is. This is the same advantage a whiteboard has, and the reason people who have never heard of any of this end up with sticky notes on a wall.
It makes the pile visible. Cards take up space. A folder does not. Seeing that you have eleven cards on one theme and two on another tells you something a search result never will.
How it differs from a Zettelkasten
They get conflated, and they are not the same system.
| Index card method | Zettelkasten | |
|---|---|---|
| Organising principle | Theme written at capture | Links between individual notes |
| Addressing | None — cards live behind dividers | Every note has an ID others point to |
| Purpose | Feed a specific project | Build a body of thought over decades |
| What you get out | An outline for the thing you are writing now | A network that suggests things you weren't looking for |
| Cost | Low. It is a shoebox | High. The linking discipline is real work |
The index card method is a project tool. Luhmann's slip-box is an accumulation tool. If you are writing a specific book, the cards are faster and the numbering scheme in folgezettel is overhead you do not need. If you want your notes from 2026 to be useful in 2034 on a topic you cannot currently name, the links are the part that pays.
The failure mode
There is one, and it kills the system: cards without sources.
You write down a striking sentence, file it, and eight months later you want to use it and cannot remember whether it is a quote, a paraphrase, or your own idea, or which of four books it came from. Now you either spend an afternoon re-reading, or you quietly leave it out. Both outcomes waste the original work.
Author, title, page number, on every card, always. It costs four seconds. It is the same discipline that makes literature notes worth keeping, and its absence is why so many reading systems produce a beautiful pile nobody can safely cite.
The second failure is milder and more common: collecting far more than you sort. Cards accumulate, the sorting session never happens, and the box becomes an ornament. That is the collector's fallacy with better handwriting, and the cure is to start a project, not to buy more cards.
Doing it digitally
The digital version keeps the constraints and loses the shoebox: one short note per idea, a single theme tag, the source in the note, and a habit of pulling a theme onto one screen and reordering it before you write. Anything that can show many short notes at once will do. What you lose is the physical spread — a table of forty cards has a bandwidth that a scrolling list does not, which is the strongest argument left for keeping this one on paper. Digital vs paper notes covers the wider trade.
Either way, the method earns its keep at the same moment: when you sit down to write and discover the material is already sorted into an argument. That is also the promise of smart notes, arrived at from a completely different direction. More formats in note-taking methods.