A Spaced Repetition Schedule for Exam Revision

How do you build a spaced repetition schedule for exams?

Start with the exam date, divide the assessed material into manageable topics, and give each topic later appointments for recall and correction. Fit those appointments into time you actually have. Use review results to decide what needs attention next, and replan missed sessions instead of automatically doubling tomorrow's workload. The calendar below is an original planning example, not a scientifically validated sequence of intervals or a promise of exam results.

What needs to go on the calendar first?

Write down the exam date, the assessed topics and your available study blocks. Include work, other deadlines and access arrangements before filling the remaining space. A plan that assumes an unavailable evening is already out of date.

Use the syllabus, assessment guidance and your teacher's instructions to identify the work. Split a heading such as “statistics” into something you can actually attempt: explaining the difference between two measures, interpreting a graph or solving a specified kind of problem. Do not treat one successful definition as evidence that a whole chapter is covered.

Give each topic a first-learning slot and later review slots. Spacing means returning to the same material across separated encounters; studying different topics on consecutive days does not, by itself, schedule a return to any one of them. For the broader method, see how to remember what you study.

Leave some calendar space unassigned. This is our practical planning choice, not a research-derived requirement. It gives you somewhere to put a topic that takes longer than expected.

Is there one best sequence of review intervals?

The studies below do not establish one universal exam timetable.

In Roediger and Karpicke's 2006 prose-passage experiments, students either restudied or took initial free-recall tests without feedback. Restudying helped more on a final test after five minutes, whereas initial testing helped more after two days or one week. These were passage-recall experiments, not a trial of the calendar below.

Kang and colleagues' 2014 vocabulary study compared expanding and equally spaced retrieval across four weeks. Final recall eight weeks after training was equivalent between schedules, although expanding retrieval produced higher average recallability during training. “Equivalent final recall” does not mean the schedules performed identically at every stage.

A 2025 online study by Rogers, Nakata and Chiu used 40 Swahili–English word pairs and a meaning-recall test ten days after the second learning session. Its adjusted analyses supported spacing over massed practice, but findings about the advantage of particular gap lengths varied with the analysis. A vocabulary task cannot specify the best timetable for your whole course.

The useful distinction is between evidence for learning methods and a particular calendar someone has designed. Our dates are editable appointments, not an optimal-interval formula.

What could a three-week plan look like?

Here is an original, fictional fragment of a revision plan. The exam is on 26 October 2026, and planning begins on 5 October: 21 elapsed days beforehand. Only two small topics are shown, not an entire course.

Topic A's gaps are 3, 6 and 8 days. Topic B's are 3, 6 and 8 days too. Each row contains four encounters, including first learning; the table therefore contains eight topic appointments. It does not specify how many minutes each appointment needs.

The final appointments are four and three days before the exam, respectively. Those gaps are consequences of this fictional calendar, not recommendations for everyone. Reserve dates of 16 and 20 October are left available for adjustments.

To adapt the table, replace the topic labels with your own assessed tasks and use dates you can keep. If the syllabus is large, check whether all topics have a place before adding extra returns to the easiest ones. If the remaining workload cannot fit, ask your teacher or course tutor to help clarify priorities; a timetable cannot manufacture more available time.

What happens during a review appointment?

An appointment needs a task, not just a topic name. Put these fields beside each date:

• Prompt: what will I answer or do without consulting the answer?
• Check: which course source or worked solution will I use afterward?
• Result: what was correct, incomplete or wrong?
• Next action: what exactly needs another attempt?

For Topic A, the prompt might be “Explain the difference between the two concepts and give one course-relevant example of each.” For Topic B, select an appropriate practice problem and identify what your course expects you to show.

After attempting the task, check it against an authoritative course source and correct the specific error. Choosing a reliable answer key matters: an unverified generated answer is not a substitute for checking the course material. Our guide to turning notes into a self-test explains how to build those prompts.

This correction step is an editorial workflow recommendation. It is not a description of the no-feedback condition in the 2006 experiment. If you need a container for the appointment, the 30-minute study session guide offers an adjustable structure, not a required dose.

What if a review goes badly?

Record the error before deciding on another date. “Forgot everything” is less useful than “defined both terms but reversed the example” or “selected the right method but made an arithmetic error.”

As a planning choice, give unresolved assessed material attention in an available slot. That may mean consulting an explanation, asking for help or making the prompt smaller before another attempt. Repeating an unclear question more often does not resolve what the question is asking.

Do not automatically widen a gap because an answer felt familiar. Equally, one unsuccessful attempt does not establish that the method is unsuitable for you. Use the actual work to revise the next task, and keep the rest of the assessed material visible.

What if you miss a scheduled review?

Change the calendar to reflect what happened. In the example, suppose Topic A's 14 October review is missed. Moving it to the available 16 October reserve date changes the gaps from 8 October to 16 October, and then to 22 October, to eight and six days.

That is a bookkeeping example, not evidence that eight and six days are the best replacement intervals. Decide whether the 22 October appointment still fits after seeing the 16 October work. If there is no reserve slot, reconsider priorities and the size of the tasks instead of silently adding an impossible catch-up block.

For your own plan, keep one short change record: original date, actual date, unfinished task and next available appointment. It distinguishes a missed appointment from material that has never been studied.

The calendar's job is to make the next useful action visible. It should remain a working document right up to the exam, with changes based on the course requirements, your available time and what your attempts actually show.

Sources

• Roediger and Karpicke, Test-Enhanced Learning, 2006: publisher abstract.
• Kang and colleagues, Retrieval practice over the long term, 2014: publisher abstract.
• Rogers, Nakata and Chiu, Optimizing distributed practice online, published online 8 January 2025: open-access research article.