I Ran a Zettelkasten for Eight Months and Kept Two of Its Rules

Zettelkasten arrives with a set of commitments: one idea per note, written in your own words, densely linked, no folders, and — in the strict readings — no hierarchy at all, so that structure emerges from the links rather than being imposed. I followed this properly for eight months and produced about 400 notes. Two of the rules changed how I think. The rest I have dropped, and I want to explain which and why without the usual defensiveness on either side.

The rule that mattered most is one idea per note. This sounds like a filing convention and it is actually a thinking constraint. When you cannot put two ideas in one note, you are forced to notice that you had two, and about a third of the time it turns out one of them was doing all the work and the other was a restatement. I catch that now while writing rather than three months later, and I have not found another practice that does the same thing.

The second rule I kept is writing in my own words with the source note separate. The discipline is that a note containing a quotation is a different kind of object from a note containing a claim, and mixing them lets you accumulate things you have not actually understood. When I enforced the separation, my volume dropped by about half immediately, which was uncomfortable and correct.

What I dropped first was the prohibition on folders. In theory links make hierarchy unnecessary; in practice, at 400 notes, I could not find anything without either remembering a specific phrase or traversing links from a hub, and traversal is slow. I now keep a shallow folder structure — six folders, one level — purely as an entry point, and it costs nothing and it makes the collection browsable by someone who is not me, which turned out to matter when a colleague asked to look at my notes on a shared project.

The bigger thing I dropped is the ambition. The literature around this method promises that with enough densely linked notes, the collection becomes a thinking partner — you follow links and find connections you did not make deliberately, and writing becomes assembly rather than composition. In eight months and 400 notes, that happened to me three times. All three were pleasant. None of them was worth the several hours a week the linking discipline cost, and I suspect the promise is calibrated to a research career rather than a job with deliverables.

What I do now: one idea per note, my own words, source separated, six folders, and links only where I would actually want to follow them — which is far fewer than the method wants, maybe one in five notes has any. That is roughly forty minutes a week rather than three hours, and my retrieval is better than it was at 400 notes, because a shallow folder plus good search beats link traversal for the way I actually look things up, which is with a half-remembered phrase and a deadline.

I would still recommend trying the strict version for a couple of months, and I would recommend it specifically for the writing discipline rather than the network. The habit of splitting a muddled note into two clear ones is the thing that transfers to everything else, including how I write emails and how I brief people. The emergent-structure part may well work for those doing sustained original research; I am not, and it is worth being honest about which of those you are before committing a year to a method designed for the other one.