The Cornell method, and the one step almost everyone skips
What the Cornell method actually is
Split your page into three areas. A narrow column down the left, a wide column on the right, and a strip across the bottom.
Questions and keywords. Written after class, not during. About 2.5 inches wide.
What was said. Short lines, indents for sub-points, gaps between ideas so you can add to it later.
Two or three sentences, in your own words, written from memory once the page is full.
That is the layout. It takes ten seconds to draw and you can stop reading most articles about Cornell notes right here, because they stop here too.
The five R's, and which one people drop
Pauk described five stages. Nearly everyone does one and four, and quietly skips the rest.
- Record. Take notes in the right-hand column during the lecture. Keep lines short. Leave white space.
- Reduce. Within a day, go back and write cues in the left column. Not summaries. Questions that the notes answer.
- Recite. Cover the notes column. Read a cue. Say the answer out loud, from memory, in full sentences. This is the step people skip.
- Reflect. Ask how this connects to last week, or to something you already believe. Write that in the summary strip.
- Review. Ten minutes a week, cues only.
Recite is the whole point. Covering your notes and forcing recall is retrieval practice, and decades of research in cognitive psychology put it well ahead of re-reading for long-term retention. The narrow column exists to make that possible. If you never cover it up, you have drawn a nice layout and learned nothing extra.
A worked example
Say you are in a lecture on supply and demand. Your notes column ends up with something like "price ceiling below equilibrium leads to shortage, queues, black markets - e.g. rent control".
Then, the next morning, you write the cue. A bad cue and a good cue look like this:
"Price ceilings"
A label. You can read it and feel like you know the answer without producing anything.
"What happens when a price ceiling sits below equilibrium, and why?"
A question you can get wrong. That is what makes it useful in a week.
The test for a cue is simple. If you cannot get it wrong, it is not a cue. Rewrite it as a question.
When it works, and when it does not
Cornell notes are good for lectures and reading where the material is mostly verbal, arrives at a steady pace, and has a shape you can question later. History, biology, law, economics, literature.
It is a poor fit in three cases, and it is worth being honest about them.
- Worked problems. A three-line derivation does not reduce to one cue, and it will not fit the narrow column. Use plain paper for problem sets.
- Fast, dense lectures. If you are struggling to keep up, splitting your attention across a layout makes it worse. Get the notes down, add the structure afterwards.
- Diagram-heavy material. Anatomy, circuits, org charts. The layout fights you. Sketch first, add cues around the edge.
Doing it on a screen
Paper is still the easiest way to run this method. The problem is that most reading now arrives as a PDF: lecture slides, a paper, a chapter scan. That is where the layout falls apart.
Typical workarounds are all bad in the same way. Annotating on top of the slide gives you no room, so notes end up crammed over the text they refer to. Keeping a separate document breaks the link between the note and the line that caused it. Printing works, and it is telling that so many people still print purely to get margins.
The other option is to widen the page itself. That is what we built SidePDF for: it adds real margin space to either side of a PDF page, so you get a narrow cue column on the left, the original slide in the middle, and a wide notes column on the right. The margins become part of the saved file, so the notes open normally in Preview or Acrobat later.
A quick setup that works
- Left margin narrow, roughly a third of the page width. Cues only.
- Right margin wide. This is where the actual notes go.
- Leave the bottom of the last page clear for the summary.
- Do the cues the next morning, not during the lecture.
Frequently asked questions
What are the five R's of the Cornell method?+
Record, Reduce, Recite, Reflect, Review. You record notes during class, reduce them to cues afterwards, recite the material with the notes covered, reflect on how it connects to what you already know, then review briefly each week.
How wide should the cue column be?+
About 2.5 inches, or 6 centimetres, on a standard page. Wide enough for a question or a keyword, narrow enough that you cannot cheat by writing full sentences in it.
Is the Cornell method good for maths and science?+
It works for the concepts and definitions. It works badly for worked problems, because a long derivation does not fit the narrow layout and does not reduce to a single cue. Most people use Cornell for theory and plain paper for problem sets.
Can I use the Cornell method on a PDF or slide deck?+
Yes, but not by typing on top of the slides. You need margin space beside the content. Printing works, or a tool that widens the page so you get a real cue column and notes column around the original.
How long does it take before it pays off?+
The recording part costs you nothing extra. The gain comes from the recite step, and that only shows up after two or three review cycles. If you only ever write notes and never cover them up, you are doing the layout without the method.
The short version
Draw the three boxes. Take notes on the right. The next day, write questions on the left, cover the right, and answer them out loud. Weekly, ten minutes, cues only.
If you only change one thing, make it the covering up. That is where the method stops being stationery and starts being studying.
Cornell notes on any PDF
Widen the margins of a lecture slide or paper and get a real cue column, without printing it.
See how SidePDF works