Zettelkasten Basics: Turn Linked Notes Into Useful Ideas

Zettelkasten is a method of developing ideas through connected notes: write each useful thought clearly, preserve its source, and link it to related thoughts with an explanation. The basics are writing, connecting, and revisiting notes so you can use them in an argument, a design, or a piece of work.

You can start on paper or in a note app. When those ideas lead to active project work, fenbs can keep the accepted decisions, lessons, and next steps available to your connected AI assistants. Your research notes remain a place to think; the project record says what you actually chose to do.

What makes a note useful later?

A useful note makes sense after you have forgotten the reading session. Give it a title that states the idea, a short explanation in your own words, and a reference precise enough to find the supporting passage. Include the limits of the claim when they matter.

For example, “Form design” is a topic. “Explain why a form asks for sensitive information” is an idea you can develop, question, and apply. Add what prompted the thought and whether it came from a source, your observation, or an untested guess. Those origins should remain distinguishable.

In his own account of working with slip boxes↗ (opens in a new tab), Niklas Luhmann describes stable note references, connections, and an index for rediscovery. You do not need to reproduce his physical numbering system. You need references that keep working as your collection grows.

Which kinds of notes should you keep?

Use a few simple distinctions rather than a complicated filing system. A quick capture is a reminder to think about something. A source note records what a particular author or document says. An idea note develops a thought you want to reuse, with references back to its evidence.

• Capture: a question, observation, or passage you do not want to lose. Review it before assuming it belongs permanently.
• Source note: the document title, author or organization, location, and a faithful account of the relevant point.
• Idea note: one understandable argument or explanation, including its uncertainty and connections.
• Entry note: a short route into a subject, pointing to a few useful idea notes and the question they address.

These are jobs a note can do, not mandatory folders or software features. A short collection may need only source notes and idea notes. The broader AI second brain guide explains how reusable knowledge can support work beyond a single writing project.

How do you start with one reading session?

1. Choose a question you care about answering, such as why people abandon an application form. Read one relevant source with that question in mind.
2. Capture the source details and the location of the useful passage. Mark exact quotations so you cannot mistake them for your own wording later.
3. Write one idea note without copying the passage. Explain the claim, its support, and the circumstances where it might not apply.
4. Find one related note. Explain whether the new idea supports it, challenges it, gives an example, or introduces a condition.
5. Use the connection in a paragraph, a discussion question, or a project proposal. Record what you still need to find out.

Suppose a design student reads a usability report about lengthy forms. Their idea note might say that a request needs both a purpose and a clear explanation. A second note might concern trust. The connection is a question about whether explaining the purpose changes the experience, rather than a claim that every shorter form is better.

Keep the example tentative until its evidence supports a firmer conclusion. A source about one study population does not automatically describe every audience. Recording that boundary now saves you from overstating the point when you return to write.

How many links should each note have?

Add the links you can explain. One meaningful connection is more useful than a list of vaguely similar titles. Write a sentence such as “This challenges my earlier assumption because the participants had already used the service.” The sentence carries the reasoning that a bare link leaves out.

Andy Matuschak describes the tradeoff in keeping notes focused on one idea↗ (opens in a new tab): notes need enough substance to be useful without becoming so broad that connections lose their meaning. His companion note on linking evergreen notes↗ (opens in a new tab) emphasizes thinking through the relationship as you make it.

Do not split a complete explanation into fragments just to increase the note count. If you need a definition, an example, and a limitation to understand the thought, keep them together. Split the note when it starts making two arguments that you would want to use independently.

Where can AI help without replacing your thinking?

An assistant can suggest clearer titles, ask questions about an argument, or compare a small selection of notes. Give it the relevant text and ask it to distinguish what the sources state from what it infers. Review the suggested connections against the original passages.

Try this request: “For these three notes, suggest one supporting connection and one possible disagreement. Identify the sentences you used. If the evidence is missing, say what I should check.” A useful answer points you back to material you can inspect, including material that complicates your first interpretation.

AI-assisted capture is a separate workflow; the guide to taking notes with ChatGPT covers that question. Here, the important habit is to accept a note because you understand and checked it, rather than because its wording sounds confident.

When does a research idea become project context?

A research idea becomes project context when it changes how you understand or pursue a particular goal. It becomes a decision only when a person chooses an action. “Explanations may improve trust” can remain an idea note while “We will explain each sensitive field in this form” becomes a project decision.

Save the source link, the reason for the choice, and the question still awaiting evidence. If testing later reveals a mistake, record the lesson and what should change next time. The fenbs workflow keeps those project records separate from an assistant’s suggestions.

How to do this with fenbs

1. Create a free account, then add the project your linked notes are helping you complete.
2. Add a reviewed context note explaining the useful finding, its source, and its limitations. Keep the longer research collection in your note tool.
3. Record the choices a person made in Decisions and rules. Add a lesson when applying an idea changes what you would do next time.
4. Connect a supported assistant through Integrations and ask it to read the project context before working.
5. Save progress before stopping, including what you checked, which question remains open, and where the next session should begin.

fenbs is free for one person, with no credit card. It does not replace a linked-note editor or automatically import your notebooks. If your only goal is private reading and idea development, the notes themselves may be all you need.

How can you tell whether the habit is helping?

Return to yesterday’s note and try to use it without reopening everything you read. Can you explain the idea, locate its evidence, follow one useful connection, and name a remaining question? Repair the missing piece instead of adding more organizational layers.

Review captures regularly and discard those that no longer matter. Revisit a connection when a new source contradicts it. A small collection that helps you write and make careful choices is a better result than a large collection you rarely open. Explore the second brain topic hub when you need another workflow for a different job.