Knowledge Concepts
What is atomic notes? A clear explanation of the one-idea-per-note principle, why it makes knowledge systems more useful, and how to write notes that actually compound.
Atomic notes are notes that capture exactly one idea — stated clearly and completely in the note itself, without requiring context from other notes to make sense — so each note can be linked to any other note it relates to, regardless of where those ideas originally came from, enabling knowledge to combine across sources and topics.
The term comes from the analogy to atoms: the smallest indivisible unit. In PKM, the atom is a single, complete idea.
The atomic note principle is most closely associated with Niklas Luhmann's Zettelkasten system and with Andy Matuschak's concept of evergreen notes. Both independently arrived at the same insight: notes organized by source (book summaries, meeting notes, article digests) create knowledge silos. Notes organized by idea, linked across sources, create compounding knowledge.
Luhmann's original Zettelkasten cards were physically small — index cards — which enforced brevity and atomicity by necessity. Each card held one idea. Each card got an identifier and was placed near related cards with explicit connections.
In the digital era, the constraint is no longer physical. Notes can be any length. This removes the natural enforcer of atomicity — which is why the concept needs to be stated explicitly in modern PKM practice.
The case for atomic notes rests on how knowledge systems actually create value: through unexpected connections.
If a note on loss aversion is buried inside a 3,000-word "Notes from Kahneman" document, you can only link to the whole document — not to the loss aversion idea specifically. When you're later writing about pricing strategy, you'd need to remember that Kahneman is relevant, find the document, scan to find the section, and extract the idea.
If "loss aversion causes people to prefer options that avoid losses over options that produce equivalent gains" is its own atomic note, you can link directly from a note on pricing strategy to that specific idea — and discover the connection when browsing either note.
Atomic notes make ideas linkable. Links make knowledge compound. Compound knowledge is the point.
Identify the single idea. When reading a source, ask: "What is the one thing here I want to remember permanently?" Not "What are the five interesting things in this chapter?" — just the one most important. You can write multiple notes from one source, but each note captures one idea.
Write the idea as a complete sentence. The note title should be a proposition, not a topic. "Atomic notes" is a topic. "Atomic notes are valuable because they enable cross-source linking" is a proposition. The difference: you can evaluate a proposition for truth. You can't evaluate a topic.
Write the body to stand alone. Include the mechanism, not just the claim. Include one concrete example. Write as if the reader has never seen this idea before and has no access to your other notes. No "as discussed above." No "see Smith."
Link to related notes immediately. After writing the note, ask: what other notes does this idea connect to? Add those links. Linked atomic notes are exponentially more useful than isolated ones.
Keep it short. Atomic notes are typically 50–200 words. If you find yourself writing 500+ words, you probably have multiple ideas competing for one note.
| Non-atomic (avoid) | Atomic (better) |
|---|---|
| "Notes from Kahneman Ch. 3" | "Loss aversion: people weight losses ~2x more than equivalent gains" |
| "Product pricing ideas" | "Anchoring effects make the first price seen disproportionately influential" |
| "Things I learned about memory" | "Sleep consolidates recent memories by replaying them in the hippocampus" |
| "SEO strategies" | "Long-tail keywords have lower competition but signal higher buying intent" |
The non-atomic notes are containers. The atomic notes are ideas. Containers can't be linked — they're too broad and their contents are invisible. Ideas can be linked precisely.
A product manager at a SaaS company:
Old approach: Creates a note "Pricing Strategy Research" and dumps everything into it — summaries of articles, ideas from books, competitor pricing observations, thoughts from a podcast. The document grows to 8,000 words. She can keyword-search it but can never link to specific ideas within it.
Atomic approach:
Six months later: She's redesigning the pricing page. Searches for "pricing" and gets a list of specific, actionable, linked atomic notes. She can pull the decoy pricing note, the anchoring note, and the annual plan note directly into a document. Total time: 8 minutes.
With the big document: she'd re-read 8,000 words, extract the relevant sections, and spend 40 minutes doing work she'd already done once.
"Atomic means short." Atomic means one idea — which usually means short, but the constraint is scope, not length. A 300-word note that develops one idea completely is better than a 50-word note that mentions three ideas superficially.
"One idea per note means I'll have too many notes." Yes — atomic note systems tend to be large. Luhmann had 90,000 cards. That's the point. A large number of well-linked atomic notes is more valuable than a small number of comprehensive documents, because the value comes from connections, and connections are only useful when they link to specific ideas.
"This is just about organization." Atomic notes are about enabling connections, not organizing for tidiness. A perfectly organized filing system of non-atomic notes provides retrieval. A well-connected system of atomic notes provides discovery of unexpected relationships — which is where new insight comes from.
"Existing notes can't be made atomic." They can, retrospectively. When you revisit an existing large note, extract the key ideas as separate atomic notes and link them back to the original. You don't need to process everything at once — atomic-ify opportunistically.
For researchers who use web content as their primary source material:
WebSnips doesn't replace atomic notes — it provides the research library that feeds them. The synthesis and extraction work is yours; WebSnips gives you the sources to work from.
What tools are best for atomic notes?
Obsidian (free, local-first, excellent [[wikilinks]]) and Roam Research (subscription, block references) are the most commonly used. Logseq is a strong free Roam alternative. The key requirement: easy linking between notes. Notion works but linking requires more friction. Plain markdown files with [[links]] in Obsidian are the minimum viable setup.
How granular should "one idea" be? Granular enough to be linkable precisely. "Behavioral economics" is too broad. "Loss aversion" is a topic. "Loss aversion causes people to prefer avoiding losses over acquiring equivalent gains" is an idea. "Loss aversion explains why free trials need to invoke the endowment effect to improve conversion" is also one idea (a more specific application). Both are valid atomic notes; the application note links to the abstract principle note.
Can I take atomic notes from books without reading the whole book? Yes — atomic notes don't require reading sequentially. You can extract notes from a chapter, an introduction, or a section. The goal is the idea, not coverage of the source.
How do atomic notes connect to writing? When you sit down to write — an article, a report, a memo — you search your atomic notes for relevant ideas. The structure of your writing often emerges from how the linked notes cluster. This is why Zettelkasten users report that writing feels "fast" after the system is built: the thinking was done in the note-writing; the writing just assembles it.
Atomic notes — one idea per note, written to stand alone, linked to related ideas — are the building block of a knowledge system that compounds over time rather than accumulating unused.
The shift to atomic notes is counterintuitive at first: you're writing more notes to capture less per note. But the return is a system where every idea can be linked precisely, where unexpected connections surface naturally, and where writing from your knowledge base takes minutes rather than hours of re-reading.
Write one atomic note today. Find the one clearest idea from something you've read recently. Write it as a complete sentence. You've started.
More WebSnips articles that pair well with this topic.
What is citation management? A clear explanation for researchers and academics — how citation managers work, the best tools, and how to avoid the most common mistakes.
What is tacit knowledge? A clear explanation for team leads and ops people — what tacit knowledge is, why it's the hardest knowledge to transfer, and practical methods to make it explicit.
What is a commonplace book? A clear explanation of the centuries-old knowledge collection practice, famous examples, and how to build a modern digital version for writers and researchers.