Work Club Join with a code Start a room

Spaced Repetition: The Schedule and How to Use It

An open monthly desk calendar on a wooden desk beside a stoneware vase, three pens and a keyboard
Photo: Swello on Unsplash

Short answer

Spaced repetition is reviewing material several times with gaps between the reviews, instead of all at once. The best gap depends on how long you need to remember: research by Cepeda and colleagues puts it at roughly 10 to 20% of the time until the test, so about a day for a test next week and three to four weeks for a test in a year. No fixed schedule such as 1-3-7-14-30 has been shown to be best, but any spaced schedule beats cramming.

Key takeaways

  • A 2006 meta-analysis found 839 assessments of spaced against massed practice in 317 experiments, and the gap that worked best grew as the test got further away.
  • In a 2008 study of 1,354 people, the best single review gap was about 3 days for a test in a week, 8 days for 5 weeks, 12 days for 10 weeks and 27 days for 50 weeks.
  • Reviewing too late costs much less than reviewing too early, so when in doubt, wait longer.
  • Ebbinghaus's forgetting curve comes from one man relearning lists of nonsense syllables, and it measures time saved on relearning, not the percentage of a lecture you forget.
  • Expanding intervals are convenient, but a 2007 experiment found equally spaced reviews did better two days later.
  • In Kornell's 2009 flashcard experiments, spacing beat massing for 90% of participants, and 72% believed the opposite.

Spaced repetition is the practice of reviewing something several times with gaps in between, and letting the gaps grow, instead of going over it repeatedly in one sitting. It works because a review after some forgetting does more for long-term memory than a review while the material is still fresh. The gap to aim for depends on your deadline: roughly 10 to 20% of the time until the test, which is about a day for a test next week and a few weeks for one a year away.

The best gap before one review, by how far away the test is1,354 people learned 32 facts, reviewed them once after a gap, and were tested later. The further the test, the longer the best gap.
1 week3 days5 weeks8 days10 weeks12 days50 weeks27 days 1 week3 days5 weeks8 days10 weeks12 days50 weeks27 days
Show the numbers
Time from the review to the testBest gap between learning and review
1 week3 days
5 weeks8 days
10 weeks12 days
50 weeks27 days

Source: Cepeda et al., 2008 (interpolated estimates for recall)

If you only want dates, skip to the schedules or let the spaced repetition schedule work them out from the day you learn a topic to the day of the exam.

What is spaced repetition?

Three terms are used almost interchangeably, and they are not quite the same thing.

Researchers also call it distributed practice. When Dunlosky and colleagues rated ten study techniques in 2013, distributed practice was one of only two given the top rating. The other was practice testing, and the two are meant to be combined: what you repeat at each spaced session should be an attempt to recall, not another read-through. That half of the method is covered in active recall, which starts with turning your notes into questions (see how to take notes). Both sit at the top of the ranking in how to study effectively.

The evidence base is large. A 2006 meta-analysis by Nicholas Cepeda and colleagues found 839 assessments of distributed practice in 317 experiments. Its main conclusion was that the gap and the deadline work together: the gap producing the best retention got longer as the time until the final test got longer.

317

experiments on spaced against massed practice in the 2006 meta-analysis.

Cepeda et al., 2006
1,354

people in the 2008 study that mapped the best gap for tests up to a year away.

Cepeda et al., 2008
90%

of participants learned more from spaced flashcards than massed ones.

Kornell, 2009

What did Ebbinghaus's forgetting curve really show?

Most explanations of spaced repetition open with the forgetting curve and a line such as "you forget 70% of what you learn within 24 hours". The curve is real. The sentence is a misreading of it.

Hermann Ebbinghaus ran the experiments on himself in 1879 and 1880. He memorised lists of nonsense syllables until he could recite them twice without error, waited between 20 minutes and 31 days, then relearned the same lists and timed how much quicker it was. His measure was the saving: the share of the original learning time he did not need the second time. In his own summary, after 24 hours "about one third was always remembered; after 6 days about one fourth, and after a whole month fully one fifth of the first work persisted in effect".

A 2015 replication, in which one person spent about 70 hours repeating the procedure, tabulates the original figures next to its own:

Ebbinghaus's forgetting curve: time saved when relearning a listThe x axis is not to scale. Each curve is one person learning nonsense syllables, so read the shape, not the exact values.
Ebbinghaus, 1880Replication, 2015
017.53552.570%20 min1 hour9 hours1 day2 days6 days31 daysEbbinghaus, 1880Replication, 2015 017.53552.570%20 min9 hours2 days31 days
Show the numbers
Time since first learningEbbinghaus, 1880Replication, 2015
20 min58.2%47.2%
1 hour44.2%37.3%
9 hours35.8%27.6%
1 day33.7%31.7%
2 days27.8%23%
6 days25.4%16.8%
31 days21.1%4.1%

Source: Murre and Dros, 2015, Table 3

The shape holds up: a steep drop in the first hour, then a long, slow tail. What does not follow is the popular percentage.

The useful lesson is the one Ebbinghaus drew: a list relearned after a gap is relearned faster, and stays longer. That is the saving that spaced repetition collects on purpose.

How far apart should the reviews be?

The best data comes from a 2008 study by Cepeda and colleagues, charted at the top of this page. People learned 32 obscure facts, reviewed them once after a gap of up to three and a half months, and took a final test up to a year after that. Three findings matter for planning.

  1. The best gap grows with the deadline. Estimated from their data, it was about 3 days for a test a week after the review, 8 days for five weeks, 12 days for ten weeks and 27 days for fifty weeks. (Of the gaps they actually tested, one day did best for the one-week test.)
  2. As a share of the deadline, it shrinks. Those gaps are 43%, 23%, 17% and 8% of the time to the test. Dunlosky's review condenses this into a rule of thumb: a gap of about 10 to 20% of the time you want to remember for.
  3. Too long is safer than too short. Recall rose steeply as the gap went from zero to the best value and then fell away gradually. The authors note that the costs of too long a gap "are much smaller than the costs of using too short a gap".

Getting the gap right was worth a lot. Against reviewing straight away, the best gap raised final recall by 10% for the one-week test, 59% for five weeks, 111% for ten weeks and 77% for fifty weeks.

To put it simply, if you want to know the optimal distribution of your study time, you need to decide how long you wish to remember something.

Nicholas Cepeda and colleagues, Spacing Effects in Learning (2008)

One caution: that study tested a single review. Nobody has mapped the ideal timing for five or six reviews in the same detail, so every multi-review schedule, including the ones below, is an informed convention.

Robert Bjork, a memory researcher who has published on spacing with Nate Kornell, explains the effect in five minutes here.

robert bjork - spacing improves long-term retention, gocognitive (5 min). Robert Bjork explains why spreading learning over time improves long-term retention and why cramming is forgotten quickly.

A spaced repetition schedule for your deadline

Count the days from first learning a topic to the exam (the exam countdown does this for several exams at once), then pick the nearest row. Day 0 is the day you learn it.

Time until the exam Review on these days Gap rule used
1 week 1, 3, 6 About a day, then a little longer
2 weeks 1, 4, 8, 13 Two to four days
1 month 1, 5, 12, 21, 28 Three to seven days, widening
3 months 2, 10, 25, 50, 80 One to four weeks, widening
No exam, long-term 1, 7, 30, then every 2 to 3 months Widening until it sticks

These are worked examples of the 10 to 20% rule, not tested prescriptions. Three notes on using them:

  • The last review sits close to the exam on purpose. A final pass the day or two before is the one useful part of cramming.
  • Each review is a recall attempt. Close the notes, write or answer, then check. Budget 5 to 15 minutes per topic, less each time.
  • Anything you get wrong moves back. Review it again within a day or two before rejoining the schedule.
Review tracker to paste into your notes
Topic:
First learned (day 0):
Exam date:
Review 1 (date):        done [ ]   missed items:
Review 2 (date):        done [ ]   missed items:
Review 3 (date):        done [ ]   missed items:
Review 4 (date):        done [ ]   missed items:
Final review (1 to 2 days before exam):   done [ ]

With several subjects, the reviews pile up quickly. Use the planner below to see how many rounds a day the whole load needs, then give the first round of each day to reviews and the rest to new material.

Fit the reviews into the days you have

Enter your hours of material and days until the exam. Count review time as part of the hours.

For a full timetable that places each subject on a date, use the study schedule maker, and see how to make a study schedule for keeping to it. The exam-season plan is in how to study for exams.

Read nextHow to Study for Exams: A Plan for 7 Days or 1A day-by-day revision plan that builds these review dates into the last week before an exam.

What about 1-3-7-14-30 and 2-3-5-7?

Both are popular, and both are sensible expanding schedules. Neither comes from an experiment that tested those numbers against others, and the guides that recommend them rarely say so. The 1-3-7-14-30 pattern suits material you want for a month or more. The 2-3-5-7 pattern, usually counted back from the exam, suits a deadline about two weeks away. If your dates do not fit either, change them; the research supports the principle, not the digits.

Do the gaps have to expand?

Not necessarily. Expanding gaps are popular because they keep the workload falling. But a 2007 experiment comparing expanding with equally spaced recall of vocabulary found that expanding helped ten minutes later while equal spacing gave better retention two days later. The authors concluded that what mattered was making the first recall hard by delaying it, not the pattern after that. In practice: do not do your first review five minutes after learning something. Wait until at least later that day, or the next day.

The spaced repetition method without an app

A plain calendar

Write the review dates for each topic in whatever calendar you already use. This is enough for most courses, where you are spacing topics (20 to 40 of them), not thousands of facts.

An orange highlighter resting on a paper wall calendar
A paper calendar and a marker are a complete spaced repetition system for a term's worth of topics. Photo: Eliza Diamond on Unsplash

The Leitner box

The paper system for flashcards is the Leitner box. According to the Anki manual, Sebastian Leitner popularised it in 1972: cards move between boxes after each review, which gives a rough measure of how well you know each one. A common three-box version:

  1. Box 1: every day. All new cards start here.
  2. Box 2: every three days. A card you get right in box 1 moves here.
  3. Box 3: once a week. A card you get right in box 2 moves here.
  4. Any card you get wrong goes back to box 1, whichever box it was in.
Read nextActive Recall: The Study Method, With ExamplesWhat to do at each review: the recall methods that make a spaced session count.

Spaced repetition flashcards: what the research says

Flashcards are where most people meet spacing, and where they most often undo it. In Nate Kornell's 2009 experiments, students learned vocabulary of the kind found in graduate entrance tests either as one large stack, so each card came round after many others, or as four small stacks studied one at a time.

Flashcard words recalled: one large stack against four small stacksThree separate experiments. Splitting cards into small piles feels efficient and reduces the spacing between repeats.
Spaced (one large stack, or spread over days)Massed (small stacks, or all in one day)
1: within one session49%36%2: across four days54%21%3: across days, with a review65%34% 1: within one session49%36%2: across four days54%21%3: across days, with a review65%34%
Show the numbers
ExperimentSpaced (one large stack, or spread over days)Massed (small stacks, or all in one day)
1: within one session49%36%
2: across four days54%21%
3: across days, with a review65%34%

Source: Kornell, 2009

Across the three experiments, spacing was better for 90% of participants. After the first session, 72% of those who thought there was a difference believed massing had worked better. Cramming a small set on the last day (34% recalled) also lost to spacing (54%) in the second experiment.

Three rules for cards follow:

  • Keep the stack large. Do not split 40 cards into four piles of ten.
  • Spread the same deck over several days instead of finishing it in one.
  • Answer before you flip, and do not retire a card after one correct answer.

If the cards are for arbitrary lists, pair them with the techniques in how to memorise fast.

Spaced repetition apps and systems

A spaced repetition system (SRS) automates the Leitner idea: you grade each card, and the software sets its next date.

  • SM-2, the original recipe. Piotr Wozniak's SuperMemo 2 algorithm, used in SuperMemo from 1987, sets the first interval at 1 day and the second at 6 days, then multiplies each interval by an "easiness" factor that starts at 2.5 and moves with your answers. Fail a card and it starts again from 1 day.
  • Anki. The best-known free option. Its manual says its scheduler was based on SM-2 and that a newer algorithm, FSRS, has been added as an alternative. With FSRS you choose a "desired retention"; the default is 90%, and the manual warns that workload climbs very quickly above that.
  • Our date generator. For topics instead of cards, the spaced repetition schedule gives you review dates between the day you learn something and the exam, in the browser.

An app is worth it for large sets of small facts: vocabulary, anatomy, drug names, case names. For understanding a method or writing essays, spaced practice questions do more than cards.

Set up spacing for one subject

Ticks are saved in this browser.

Why spacing feels worse than it is

By the time a spaced review comes round you have forgotten some of the material, so the session feels harder and less successful than a cram. That is why most of Kornell's participants walked away believing massing had served them better when it had not. The difficulty is the method working. It is also why sticking to the dates takes some structure: a fixed daily slot, a visible calendar, or other people.

On the days the reviews feel like a chore, how to stay motivated to study and how to stop procrastinating are the guides to read. Reviews are short, so they fit well into timed rounds: one 25-minute block, as described in the Pomodoro method for studying, usually clears a day's reviews. If you are unsure where in the day to put that block, the best time to study looks at morning against evening, and how many hours you should study helps you size the rest of the day around it.

Keeping a review habit with other people

A standing daily session with friends is one way to make sure reviews happen. Agree a time, start with one round of reviews each, then move on to new work. Our guide to studying with friends online has formats for this, and group study covers quizzing each other on the cards.

In Work Club, a group shares one timer, with 25-minute focus rounds and 5-minute breaks by default. A simple routine is to make the first round of every session "reviews only" for everyone. If you make a free account, your streak follows you across devices, which gives the daily habit something to hang on.

One shared countdown. Make the first round of the day the review round. Try it in a room

Common questions

What is the best spaced repetition schedule?

There is no single best one, because the right gap depends on how long you need to remember. A sound rule from the research is to make the gap about 10 to 20% of the time until the test. For an exam in a month, that means reviews every three to six days; for an exam next week, daily.

Is the 1-3-7-14-30 schedule backed by research?

Not specifically. It is a reasonable expanding schedule, and spacing in general is very well supported, but those particular numbers do not come from an experiment. The 2-3-5-7 method is the same kind of convention. Use them as defaults and adjust to your exam date.

Does spaced repetition only work with flashcards?

No. Flashcards are the common form because they are easy to schedule, but the spacing effect has been found with many kinds of material, not only word lists. You can space practice questions, essay plans or blank-page recall of a topic in the same way.

How many times should I review something?

Plan for at least three reviews after first learning it, and keep reviewing anything you still get wrong. In flashcard apps the number is decided for you: a card you keep getting right is shown less and less often.

What happens if I miss a review day?

Very little. In the 2008 study, recall fell away slowly when the gap was longer than ideal and steeply when it was shorter. Do the review when you can and carry on; do not restart the schedule.

Is cramming ever worth it?

It can raise your score on a test the next morning, and it is better than not studying. The cost is that most of it is gone soon after, which matters for cumulative courses. A spaced plan with a final review the night before gets you both.

Sources

  1. Distributed practice in verbal recall tasks: A review and quantitative synthesis, Psychological Bulletin (2006)
  2. Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention, Psychological Science (author's copy, York University) (2008)
  3. Memory: A Contribution to Experimental Psychology, Chapter 7: Retention and Obliviscence as a Function of the Time, Hermann Ebbinghaus (1885), Classics in the History of Psychology, York University (1885)
  4. Replication and Analysis of Ebbinghaus' Forgetting Curve, PLOS ONE (2015)
  5. Optimising Learning Using Flashcards: Spacing Is More Effective Than Cramming, Applied Cognitive Psychology (author's copy, Williams College) (2009)
  6. Expanding retrieval practice promotes short-term retention, but equally spaced retrieval enhances long-term retention, Journal of Experimental Psychology: Learning, Memory, and Cognition (2007)
  7. Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology, Psychological Science in the Public Interest (2013)
  8. Background: Active Recall Testing and Spaced Repetition, Anki Manual (2026)
  9. SuperMemo 2: Algorithm, SuperMemo (Piotr Wozniak) (1990)
  10. Deck Options, Anki Manual (2026)

Published 14 November 2026. Last checked and updated 14 November 2026. See how these guides are written.