Spaced repetition: the complete guide

Spaced repetition is the most reliable method anyone has found for remembering things long-term, and almost nobody uses it consistently. This guide explains what it is, why it works, and how to actually put it into practice, with links to deeper pieces on each part.

I am not a memory researcher. I am a self-taught learner who got tired of forgetting what I studied, read the research, and ended up building a tool around it. So this is a practical guide, grounded in the science but written for the person who just wants to remember more of what they learn.

What spaced repetition is

Spaced repetition is reviewing material at increasing intervals over time instead of all at once. You learn something today, review it tomorrow, again a few days later, again the next week, then the next month. Each review comes just as the memory is starting to fade, and each one pushes the next review further out.

That is the whole idea. It is not an app, an algorithm, or a deck of flashcards. Those are tools that implement it. The method itself is just a schedule of well-timed reviews, and you can run it on paper if you want.

What makes it powerful is efficiency. A handful of two-minute reviews spread across a month will plant a fact more firmly than an hour of cramming the night before, and cost you less total time. You are working with how memory actually consolidates instead of against it.

The science: the forgetting curve

Human memory has a predictable shape, and it slopes downward fast. Learn something today and, without reinforcement, most of the detail is gone within a day or two. This is the forgetting curve, first measured by Hermann Ebbinghaus in the 1880s.

The useful part is what happens when you review. Each well-timed review flattens the curve: the memory decays more slowly afterward, so the next review can wait longer. Reviews compound. The forgetting curve, explained goes deeper on the mechanism, but the practical takeaway is simple: the goal is to review just before you would have forgotten, and to let the gaps grow each time.

This effect, that spacing reviews out beats massing them together, is one of the most replicated findings in learning science. It is called the spacing effect, and it has held up across more than a century of studies. For a narrative take on the same idea (and the book that convinced me to build around it), see why spaced repetition actually works.

The other half: active recall

Spacing tells you when to review. It does not tell you how. The how matters just as much, and the answer is active recall: retrieving the answer from memory instead of rereading it.

Rereading feels productive because the material grows familiar, but familiarity is not the same as recall. The act of pulling an answer out of your head, and occasionally failing, is what actually strengthens the memory. This is the testing effect, shown most famously in work by Roediger and Karpicke. Spacing and active recall reinforce each other: the small amount of forgetting that spacing allows is exactly what makes the next retrieval effortful enough to count.

So the complete method is two rules working together: retrieve rather than review, and do it at growing intervals.

The algorithms

Most spaced-repetition tools automate the scheduling with an algorithm. You do not need to understand them to benefit, but it helps to know the landscape.

• The Leitner system is the analog ancestor: physical cards move between boxes, and each box is reviewed less often. Simple, manual, and still a fine way to learn the concept.
• SM-2 is the algorithm Piotr Woźniak designed for SuperMemo in the 1980s. It is the one Anki popularized and the one most people have used without knowing its name. You grade each card and it adjusts the next interval.
• FSRS (Free Spaced Repetition Scheduler) is the modern, machine-learning-based successor, now built into Anki and several other tools. It predicts your forgetting more accurately than SM-2 and is the current state of the art for card-level scheduling.

All of these schedule individual cards. That is the right tool when your material is genuinely flashcard-shaped. It is not the only way to apply spacing, which matters for the next section.

When spaced repetition works, and when it doesn’t

Spaced repetition is unbeatable for discrete, retrievable facts: vocabulary, definitions, formulas, dates, anatomy, names. If you can put it on a card with a clear answer, spacing will help you keep it.

It is a worse fit for things that do not compress into a question and answer: understanding a proof, building intuition for a system, practising a physical skill. You can sometimes break those into retrievable sub-facts, but forcing everything onto flashcards is a common way to waste effort. For material like that, the value of spacing is in scheduling when you revisit the source (the chapter, the notes, the project) rather than drilling cards.

That distinction, drilling cards versus scheduling sources, is the core idea behind the tool I built, and it is why a planner and a flashcard app are different things.

How to actually use it

You do not need an app to start. The method comes down to three decisions.

1. Pick clear units. “Spanish” is too vague to review. “The 200 most common Spanish verbs” or “chapter 4 of the textbook” is reviewable. The sharper the unit, the easier the review.
2. Use an expanding rhythm. A simple sequence works for almost everything: review one day later, three days later, a week, two weeks, a month, three months. The exact numbers matter less than the fact that the gaps grow.
3. Cap your daily load. This is the rule everyone skips and the reason most attempts fail. If you let reviews pile onto a single day, the first time life interrupts you face a wall and quit. A daily limit, with overflow rolled forward, is what keeps the habit alive. This is important enough that it has its own guide: spaced repetition without the burnout.

For a worked example on a real subject, see spaced repetition for language learning, which is how I learned English vocabulary.

The tools

Once you want software to run the schedule, the main options divide by philosophy:

• Anki is the powerful, free, open-source standard for flashcard drilling. See Memset vs Anki.
• Quizlet is the mainstream option with a huge shared library of ready-made sets. See Memset vs Quizlet.
• RemNote folds notes, PDFs, and flashcards into one knowledge base. See Memset vs RemNote.
• Brainscape sells expert-made certified decks with a confidence-rating drill engine. See Memset vs Brainscape.
• Memset (the tool I built) is the odd one out: it does not store or drill cards. It sits on top of whatever you already study and schedules what to revisit today, with a daily limit so the reviews never pile up.

If your material is flashcard-shaped, pick a drill tool and let it schedule cards. If your material lives in books, notes, and courses, and you keep losing track of what to revisit, a planner is the better fit. Many people use both.

Where to start

If you read one more thing, make it the piece on why short daily limits keep the habit alive, because the scheduling problem, not the science, is what defeats most people. The science is settled. The hard part is showing up for the small reviews, week after week, without burning out.

That is the entire problem Memset is built to solve. You can list your first few sources tonight and get a schedule that tells you what to revisit tomorrow, sized so you can actually keep it.