Productivity

How to Take Better Notes

Good notes are not a transcript. The point is to force decisions while you write — what matters, what connects to what, what you did not understand — because those decisions are the learning. Notes you can copy without thinking teach you nothing.

The most useful single change: leave a wide margin, and after the lecture write questions in it that your notes answer. That converts a passive record into a self-test you can use for months.

Most note-taking advice is about systems — Cornell, outlining, mind maps. Systems help, but they are downstream of one decision: are you capturing words, or processing ideas? This guide covers what the evidence supports, three methods worth knowing, and how to make notes you will actually reuse.

Why transcribing fails

Typing is faster than handwriting, which sounds like an advantage and is usually the opposite. When you can keep up with a speaker, you can transcribe without understanding — the words pass through you to the page. When you cannot keep up, you are forced to compress: decide what matters, drop the rest, put it in your own phrasing.

That compression is the work. It is also why studies comparing longhand and laptop notes tend to find longhand better for conceptual questions even when laptop users capture more words. More words, less processing.

This does not mean you must handwrite. It means that if you type, you have to impose the constraint yourself — deliberately paraphrase rather than transcribe, or you get the speed without the benefit.

Method 1: Cornell

The most useful method for lectures, because it builds retrieval practice into the format. It comes from Cornell University's Learning Strategies Center, which still publishes the original description.

Divide the page into three zones:

  • Right column (about 70%) — your notes during the session.
  • Left column (about 30%) — left empty during the lecture. Afterwards, write questions here that the notes on the right answer.
  • Bottom strip (a few lines) — a summary in your own words, written later.

The left column is the entire point. Cover the right side and you have a question bank; the notes become a test rather than something to re-read. Writing those questions afterwards is itself a review pass, and deciding what question a paragraph answers forces you to work out what it was actually about.

The bottom summary catches the case where you cannot state the main idea — which is worth discovering now rather than in an exam.

Method 2: outlining

Indentation to show hierarchy. Main point at the left, supporting detail indented under it.

Photosynthesis
  Light-dependent reactions
    - occur in thylakoid membrane
    - produce ATP and NADPH
  Calvin cycle
    - occurs in stroma
    - uses ATP/NADPH to fix CO2

Good for structured material where the speaker follows a clear hierarchy. Poor for discussion-based sessions that jump around, because you spend the session fighting the format.

Its real weakness is that it records structure without forcing you to evaluate importance — everything gets a bullet. Pair it with the Cornell question column and that gap closes.

Method 3: mapping

Concept in the middle, related ideas branching outward, lines showing relationships.

Genuinely better than linear notes for material where the connections are the content — causes and effects, systems that feed into each other, comparisons between theories. Linear notes flatten a network into a list and lose exactly the information you needed.

Weak for sequential material and for anything you need to search later. Use it as a supplement for the two or three genuinely networked topics in a course, not as your default.

Choosing

MaterialMethod
Structured lectureCornell, with an outline in the notes column
Discussion or seminarCornell — the loose format tolerates jumping around
Interconnected conceptsMapping
Procedures and stepsNumbered outline
Reading a textbookQuestions first, then notes that answer them

The habits that matter more than the method

Write less. If your notes are as long as the source, you transcribed. Aim to cut by at least half. A page of notes you understand beats five pages you copied.

Use your own words. The moment you paraphrase, you have to understand the idea well enough to restate it. Copying a definition verbatim proves nothing except that you can copy.

Mark confusion explicitly. Put a ? next to anything you did not follow. Most people quietly skip past these and never return; a visible mark turns confusion into a to-do list. This is the highest-value habit here and it costs one character.

Review within 24 hours. Ten minutes filling gaps while the session is still in memory saves an hour of reconstruction later. This is also where you write the Cornell questions.

Date and title everything. Trivial, and the reason most people cannot find anything in March that they wrote in October.

Notes are an input, not the goal

The most common failure is treating finished notes as the achievement. Beautiful, colour-coded, complete notes that are then re-read a few times before the exam produce very little learning — re-reading builds familiarity rather than recall.

Notes are raw material for retrieval practice. The sequence that works:

  1. Take notes that force processing during the session.
  2. Within a day, add questions and mark what you did not understand.
  3. From then on, use the questions — not the notes — to test yourself.
  4. Only open the notes to check answers and fill gaps.

Step 3 is the one people skip, and it is where the benefit is. Our guide to study techniques that actually work covers the evidence for testing yourself over reviewing, which is the principle this whole sequence rests on.

Digital or paper?

Both work; the trade-offs are real and different.

PaperDigital
ProcessingForced by slower writingMust be imposed deliberately
DiagramsImmediateSlow unless you have a stylus
SearchNoneExcellent
DistractionNoneSubstantial
BackupLose it and it is goneAutomatic

A reasonable compromise: handwrite during the session, then type a condensed version within a day. The retyping is a review pass rather than duplicated effort, and you end with searchable notes that were processed twice.

If you use an AI assistant to help organise or quiz you on notes, keep it to your own material — our guide to free AI tools for students covers doing that without importing fabricated facts, and writing better AI prompts covers getting useful answers out of one.

Frequently asked questions

What is the best note-taking method?

Cornell, for most lecture material, because the question column builds self-testing into the format. Mapping is better where the relationships between ideas are the point, and outlining suits highly structured content. The method matters less than whether you paraphrase rather than transcribe.

Is it better to write notes by hand or type them?

Handwriting tends to produce better conceptual understanding, most likely because you cannot write fast enough to transcribe and must therefore compress and select. Typing is fine if you deliberately paraphrase instead of capturing verbatim — the benefit comes from the processing, not the pen.

How do I stop writing down everything?

Set a rule that your notes must be at most half the length of the source, and use your own words. If you catch yourself copying a sentence, stop and write what it means instead. Marking the gaps with a question mark is more useful than filling them in during the session.

What is the Cornell note-taking system?

Divide the page into a narrow left column, a wide right column and a strip at the bottom. Take notes on the right during the session, write questions on the left afterwards, and summarise at the bottom. Covering the right column turns your notes into a self-test.

How soon should I review my notes?

Within 24 hours, for about ten minutes — fill gaps, add questions, mark what you did not understand while the session is still fresh. After that, review by testing yourself with the questions rather than re-reading the notes.

Should I rewrite my notes to make them neater?

Only if the rewriting involves condensing or reorganising, which forces processing. Copying notes out neatly is close to transcription and produces very little learning for the time it takes.

Conclusion

The test of a note-taking system is not how the notes look but whether writing them made you think, and whether you can use them to test yourself later. Cornell passes both, which is why it survives.

If you change one thing, add the question column. It takes ten minutes after each session and converts a record you would re-read into a tool you can practise with.

Comments