How to make good flashcards: rules, examples and fixes

By oosync editorial team · Updated 2026-10-14 · 6 min read

Flashcards are one of the simplest ways to put active recall and spaced repetition into practice. They are also easy to get wrong. A deck of vague, overloaded cards is tiring to review, hard to remember and gives you a false picture of what you know.

The quality of a flashcard deck depends far more on how the cards are written than on which app you use. This guide sets out the rules that matter, with before-and-after examples you can copy.

Why flashcards work, when they work

A flashcard forces you to produce an answer before you see it. That act of retrieval, repeated over spaced intervals, is what strengthens memory. Decades of research on the testing effect and on distributed practice support both halves of that process.

But a flashcard only works if it actually makes you retrieve something specific. If the question is vague, you end up recognising the answer rather than recalling it, and recognition is much weaker. Most of the rules below are ways of making sure every card asks for one clear thing.

Rule 1: one idea per card

This is the most important rule. Piotr Wozniak, who developed the SuperMemo software, called it the minimum information principle. Small cards are easier to review, easier to schedule and easier to diagnose when you get them wrong.

Before:

Q: What are the features of a contract? A: Offer, acceptance, consideration, intention to create legal relations, capacity, and certainty of terms.

You will half-remember this card forever. Some days you get four of six, some days five, and you never know which one you are missing.

After:

Q: In contract formation, what is "consideration"? A: Something of value given in exchange for a promise. Each side must give something.

Q: Name the element of contract formation that asks whether the parties meant their agreement to be legally binding. A: Intention to create legal relations.

If you genuinely need the whole list, add one more card for it, but only after the individual elements are secure.

Rule 2: make yourself produce, not recognise

A card with a yes or no answer, or an answer you can guess from the wording of the question, tests very little.

Before:

Q: Is the mitochondrion the site of aerobic respiration? A: Yes.

After:

Q: In which organelle does most aerobic respiration take place? A: The mitochondrion.

Better still, ask something that requires understanding:

Q: Why do muscle cells typically contain many mitochondria? A: They have high energy demands, and mitochondria produce most of the cell's ATP through aerobic respiration.

Rule 3: ask why and how, not only what

University exams rarely reward definitions on their own. They ask you to explain, apply and compare. Your cards should do the same. Asking yourself why something is true is a technique called elaborative interrogation, and it tends to produce deeper understanding than memorising the statement.

Rule 4: use cloze cards for terms in context

A cloze card blanks out a key term in a sentence:

In a randomised controlled trial, [...] reduces the chance that differences between groups are caused by who was assigned to which group. Answer: random allocation (randomisation)

Cloze cards are quick to make and keep the term attached to its meaning. Keep them to one blank per card, or one blank per version of the card. A sentence with four blanks becomes a guessing game.

Rule 5: give enough context to be unambiguous

If a question could have several correct answers, you will get it "wrong" for the wrong reasons. Add a short context cue.

Before:

Q: What does BP stand for?

After:

Q: (Nursing observations) What does BP stand for, and what two numbers does it report? A: Blood pressure. Systolic pressure over diastolic pressure.

A subject tag or a short label in brackets is often enough.

Rule 6: break lists into pieces

Lists are where flashcards go to die. If you have to learn a list, try these approaches:

• Card each item separately, with a cue that points to it: "First stage of Gibbs' reflective cycle?" then "Stage after Feelings?"
• Use a cloze card for each position in an ordered list.
• Learn the logic of the list, not just its order. If you understand why the stages follow each other, you need fewer cards.

Rule 7: add a picture where it helps

For anatomy, structures, diagrams, maps and graphs, an image on the front can be a better prompt than words. "Label the structure marked A" or "What happens to this curve if demand rises?" makes you retrieve something visual and verbal together. This kind of dual coding can help, provided the picture is directly relevant rather than decoration.

Rule 8: write cards in your own words

Copying a sentence from a textbook onto a card is easy. Rephrasing it forces you to understand it, which is a small but real benefit before you even start reviewing. It also makes the card shorter. If you cannot put an idea in your own words, that is a sign to go back to the source first.

Worked example: from notes to cards

Suppose your notes on a statistics lecture say:

A Type I error is rejecting the null hypothesis when it is actually true (false positive). Its probability is the significance level, alpha, often set at 0.05. A Type II error is failing to reject a false null (false negative). Increasing sample size reduces Type II error.

A good set of cards from this paragraph:

1. Q: What is a Type I error? A: Rejecting a null hypothesis that is actually true (a false positive).
2. Q: What is a Type II error? A: Failing to reject a null hypothesis that is actually false (a false negative).
3. Q: Which error rate does the significance level (alpha) set? A: The Type I error rate.
4. Q: Name one way to reduce the chance of a Type II error. A: Increase the sample size (increasing power).
5. Q: A drug trial concludes a drug works when it actually does not. Which type of error is that? A: Type I.

Notice that card 5 is an application. It tests whether you can use the idea, not just repeat it.

How to review your deck

• Say or write the answer before you flip. Thinking "oh yes, I know this" is not retrieval.
• Be strict when you mark yourself. If you got it half right, count it as wrong.
• Review a little every day. Fifteen minutes daily beats two hours once a week.
• Mix subjects or topics in some sessions. Interleaving makes you work out which idea applies, which is closer to what exams ask.
• Edit as you go. If a card annoys you, confuses you or is always easy, fix it, split it or delete it.

Signs your deck needs a clean-up

• Reviews take longer every week and you dread opening the app.
• You keep failing the same cards.
• You can answer cards but cannot use the ideas in practice questions.
• Many cards are about things that will never be examined.

When this happens, stop adding cards for a week, delete anything trivial, and rewrite your worst ten cards using the rules above.

Generated flashcards: useful, but check them

Generating cards from your material can save a lot of time, especially for long readings. In oosync, flashcards are generated from your own uploaded slides and notes, in both question and answer and cloze formats, scheduled with spaced repetition, and you can export them to Anki. If that sounds useful, see flashcards from slides.

Treat any generated card, from any tool, as a draft. Check it against your source, delete what is trivial, and rewrite anything vague. That editing pass is a short, effective study session in its own right.

Flashcard checklist

1. One idea per card.
2. The answer must be produced, not guessed from the question.
3. Include why and how cards, not only definitions.
4. Use cloze cards with a single blank.
5. Add a context cue when a question could be ambiguous.
6. Break lists into separate cards.
7. Use images for anything visual.
8. Write in your own words.
9. Review daily, mark strictly and edit as you go.