Session replays: 20 minutes of active revision instead of rewatching the whole session

The problem: watching the session again is not the same as knowing how to do it again

After a tutoring session, the temptation is understandable: since the whole explanation is available on video, why not play the session again and watch it a second time?

The problem arises when watching becomes the main activity. During the replay, everything seems familiar: the tutor explains a step, the student recognises the formula, sees why the correction works and may feel they have regained mastery of the topic. But this recognition does not answer the decisive question: without the tutor and without the video, could they rebuild the method?

The research summarised by John Dunlosky on learning strategies draws precisely this distinction between simple rereading or re-exposure and retrieving information from memory. In his summary published by the American Federation of Teachers, practice testing and distributed practice are among the most robust strategies studied. They underpin the NOEO method for turning a tutoring session into a study sheet, flashcards and a quiz. The value of testing is not only to measure what you know: trying to retrieve information also helps identify what is still missing.

What gives an impression of mastery without always proving learning

Several behaviours can seem productive because they are comfortable and give the feeling of working: watching a long sequence without a pause, copying out the tutor’s notes exactly, recognising a solution before it appears or speeding up a video you already know.

These activities are not useless in themselves. Watching an explanation again can be necessary when a step has been forgotten or misunderstood. The difficulty comes from the lack of any test after watching.

A study by Palmer, Chu and Persky compared rewatching a recorded lecture with a knowledge retrieval activity among 102 pharmacy students. A week later, the two conditions produced no significant difference on the main retention score, but the students had spent much more time rewatching: about 35 minutes against 12 minutes for retrieval. The authors concluded that retrieval could be a more time-efficient strategy in their experimental context. The study can be read in the American Journal of Pharmaceutical Education.

Caution is needed: these participants were university students working on recorded lectures, not collège or lycée students using the replay of a tutoring session. This study therefore does not show that a student should systematically replace the video with a quiz. It illustrates a more general distinction: time spent in front of a recording and the ability to retrieve and then use information are two different things.

To explore the mechanism of retrieving from memory specifically, without repeating its full definition here, see our article on active recall.

A 20-minute protocol for using the replay actively

The following protocol is not a scientifically optimal length that suits every student. Twenty minutes is a practical framework: long enough to diagnose a blockage and rework a method, short enough to avoid mechanically replaying a whole session.

Time	Action	Goal
0–3 min	Define precisely what you need to find and try to remember it without the replay	Identify what is really available in memory
3–8 min	Find only the relevant part of the session and watch it	Fix the missing point without needlessly rewatching the whole session
8–14 min	Close or pause the replay, then rebuild the method alone	Move from recognition to production
14–18 min	Do a similar exercise, answer a question or explain the reasoning out loud	Check that the method can be reused
18–20 min	Note what has been acquired and the remaining difficulty	Know what to rework next or what to ask the tutor

0 to 3 minutes: start without the video

Before opening the replay, write down the question you want to answer. It needs to be precise enough: “how do I start this exercise?”, “why do we use this theorem here?”, “which step is missing from my factorisation?” or “how had the tutor structured the introduction?”.

Then try to answer it from memory. Even an incomplete attempt is useful: it reveals the boundary between what you already know and what you need to find again.

Opening the video straight away removes this information. You risk recognising the method before you have checked whether you were able to produce it.

3 to 8 minutes: find the useful part, not the whole session

Now the replay comes in. The aim is not to go back to the first minute of the session but to find the explanation that answers the blockage you identified.

Listen for a few minutes with a question in mind. When you find the important step, stop the video. If the tutor is working through a proof or solving an exercise, it can even help to pause before the end and try to predict the next step.

Here the video plays its best role: restoring a context or an explanation that the student could no longer rebuild alone.

8 to 14 minutes: remove the help and redo

This is the most important step of the protocol. Close or hide the replay, take a sheet of paper and redo the reasoning from the beginning.

In maths, this might mean redoing the exercise without looking at the correction. In history, rebuilding the steps of a causal argument. In Philosophy, explaining a distinction in your own words. In a language, rephrasing a rule and then creating new examples.

You can go back to the video for a few seconds if you are really stuck, but each return must answer a specific question. The aim is not to copy the solution bit by bit by switching between the sheet and the screen.

14 to 18 minutes: change the task slightly

Managing to reproduce the session’s exercise exactly is not always enough to show that the method has been understood. The context is still very close to that of the replay.

So add a small variation: change the values in an exercise, answer a similar question, explain the principle without your notes or ask yourself in what other case the method could be used.

A meta-analysis published in 2025 in Educational Research Review examined 54 studies comparing different forms of activity during video learning with passive viewing. It reports positive effects on average across several learning measures, while showing that results depend in particular on the characteristics of the learners and of the videos. Most of the studies included involved adults and STEM subjects, so it would be going too far to derive a universal recipe for all students. The general signal is nonetheless consistent with this protocol: an educational video becomes more valuable when it triggers cognitive activity rather than simple viewing.

18 to 20 minutes: produce a record for the next revision session

Finish with two sentences: “now I know…” and “I’m still stuck on…”.

This last minute avoids starting the next revision session without knowing where to pick up. If the problem is solved, it can be turned into a recall question, into a flashcard when it lends itself to one, or checked later with a quiz. The corresponding methods are covered separately in our articles on flashcards and on quizzes for spotting gaps.

Example: reviewing a maths method without rewatching an hour of tutoring

Imagine that, after a session on quadratic equations, a student is faced with an exercise and knows they need to factorise, but no longer remembers how to get from the resulting expression to the solutions.

A poor strategy would be to replay the entire session and wait for the relevant part to come round again. For much of the viewing, the student would be watching things they have already mastered.

With the short protocol, they start by writing down what they remember. They attempt the calculation. They then identify their blockage precisely: they get the factorised form but hesitate over using the zero product property. They then open only the part where this point was explained.

After a few minutes, they close the replay and redo the exercise from the factorised form. Then they do a second exercise with different coefficients. Finally, they note: “zero product understood; I still need to make factorisation automatic”.

The same video has been used, but the logic changes completely. The replay is no longer the work itself: it simply provides the missing information between two of the student’s attempts.

How NOEO can support this protocol after a session

On NOEO, a replay is only available when a recording has been authorised and actually made. Without valid Replay consent, the session takes place normally without recording. When a replay has been made and can still be viewed, access is provided for 3 months.

The main benefit is to place the video back in the context of the session. After the session, the student can find the session report, the documents used and the work to prepare. These elements make it possible to choose what is worth reviewing instead of browsing the replay without a goal.

For example, an area for improvement noted in the session report can become the starting point of the protocol: attempt the skill concerned, go back to the relevant part of the session if necessary, then redo it without the video. The replay then complements the other educational records instead of replacing them.

How things work before, during and after the session is described in detail on the How does private tutoring work on NOEO? page. To place this follow-up within the wider organisation of remote support, our guide to online private tutoring complements this approach.

It is also worth bearing in mind that a video call recording involves data such as image and voice. The CNIL (the French data protection authority) sets out several precautions for video conferencing tools, notably regarding informing users and the confidentiality of exchanges.

The mistakes that make replays less useful

• Systematically replaying the whole session: the replay should answer a precise question when watching it in full is not necessary.
• Watching before trying: trying first tells you what has really been mastered.
• Confusing recognition with independence: understanding the tutor during the video does not prove you will be able to do it again alone.
• Copying the correction while watching: the work must also exist without the support.
• Never changing the exercise: a small variation helps check that the method is not tied only to the example from the session.
• Piling up replays to watch: it is better to deal with a concrete blockage than to turn every recorded session into a new list of videos to watch.

How to check that the method is working

The best indicator is not the number of minutes of replay watched. It is what the student then manages to do without the replay.

After a few uses, three questions provide a simple test: do they find the method more quickly before opening the video? Do they need to review shorter parts? Do they succeed at a similar task without help?

If the answers move in this direction, the replay becomes less and less of a crutch and more and more of an occasional check. That is precisely the aim: having access to the session when an explanation is missing, while developing the ability to work when the session is no longer in front of you.

Conversely, if the student rewatches the same part several times but is still unable to get started alone, watching more is probably not the next useful action. You need to identify the missing prerequisite, simplify the exercise, try a different explanation or go back to the tutor with a specific question.