The short version: choose one good video for one goal, stop every few minutes to answer questions from memory, explain each part in your own words, and test yourself again a day and a week later. Rewatching and copying the transcript feel productive, but the research says they do much less than recall.
The rest of this guide explains why, step by step, with the studies behind each tip.
Can you actually learn from YouTube?
Yes, with a catch. A 2021 review of 105 studies covering 7,776 students found that video helped learning in both cases it looked at, but very differently. When video replaced other teaching, the benefit was small (g = 0.28). When video was added to existing teaching, the benefit was large (g = 0.80).
That matches how most students use YouTube: as a second explanation next to a course, not as the course itself. The video is rarely the problem. What you do while and after you watch is what decides whether it sticks.
Why watching feels like learning when it isn't
A clear, confident explainer makes material feel easy. In one experiment, students who watched a fluent instructor predicted they would remember more than students who watched a hesitant one, but on the actual test the two groups scored about the same. A polished YouTuber raises your confidence, not necessarily your score.
Students also tend to pick the strategies that feel good. In a survey of 177 college students, 84% said they reread their material, and only 11% said they tested themselves. Rereading, and rewatching, feel fluent because the material is familiar. Familiar is not the same as remembered.
So the rule for everything below is simple: judge your learning by what you can recall with the video closed.
Before you press play
Pick one goal and one video. "Understand how neural networks pick an answer" is a goal; "learn machine learning" is not. Choose the video that matches the goal, check that the creator knows the subject, and skim the chapters to see what it covers.
Make a playlist your syllabus. If a topic needs several videos, put them in a playlist in the order you will watch them, and open them from there instead of the home page or the sidebar.
Turn off autoplay. Autoplay decides what you watch next. Turn it off so the end of a video is a stopping point, not a doorway to the next one.
Plan your stops. Look at the chapter markers or the length and decide where you will pause, roughly every 5 to 10 minutes. The next section explains why deciding in advance matters.
While you watch
Stop every 5 to 10 minutes and answer from memory
This is the single most useful habit, and it has direct evidence behind it for video lectures. In a 2013 study in PNAS, students watched a 21-minute statistics lecture split into four parts. One group answered a few questions after each part, one group restudied the material instead, and one group did neither.
- On the final test, the group that answered questions scored 90%, against 76% for restudying and 68% for neither.
- The tested group's minds wandered far less: about 19% of the time, against roughly 40% for the other two groups.
- They also took more notes and felt less anxious about the final test.
Later results are more mixed. A larger 2022 study of 195 students found only a small drop in mind wandering and no clear gain on the final test. The wider research on recall practice, covered below, is still the best reason to stop and answer questions.
At each stop, close the video or look away and ask yourself two or three questions: What was the main claim? Why is it true? What example did they use? Then check against the video.
Decide where to pause before you start
Being able to pause is not the same as pausing. In a study with 68 students learning a procedure, a video that paused itself after each step helped them carry it out better than a video with a pause button, and the students given the pause button barely used it. When you are absorbed, you do not stop. That is why planned stops work better than good intentions.
Watch at 1.5x or slower when the material is new
Speeding up is less harmful than people fear, up to a point. In a 2022 study with 231 students watching real YouTube lectures, 1x, 1.5x and 2x produced similar test scores; only 2.5x was clearly worse. But a 2025 meta-analysis of 12 studies found that faster playback raises mental load and lowers retention on average.
A reasonable rule: stay at or below 1.5x for new or dense material, use 2x only for content you already half know, and spend the time you save on answering questions rather than on another video.
Taking notes that actually help
A transcript is not notes
YouTube's transcript, or an AI summary of it, gives you the words. It does not give you the understanding, because you did not have to produce anything. Use transcripts to check a detail or find a timestamp, not as your study material.
Explain each part in your own words
Good evidence for this comes from a 2020 study in which 196 students watched a lesson on how kidneys work. Students who wrote an explanation after each part of the video beat students who rewatched each part, both on remembering the content and, by an even wider margin, on applying it to new problems.
So after each stop, write two or three sentences explaining the idea as if to a friend, from memory, then check them against the video.
Handwriting versus typing matters less than you have heard
A famous 2014 study suggested that laptop notes hurt learning because students type word for word. A direct replication in 2021 found that laptop notes did contain more verbatim text, but quiz scores did not differ. Use whichever you will actually review. What you do with the notes afterwards matters more than the tool you write them with.
Turn notes into questions
The most useful notes end in questions. For each section, write one or two questions you could answer next week. They become your quiz for the next step.
After the video
Test yourself instead of rewatching
Rewatching wins in the short term and loses in the long term. In a classic 2006 experiment, students who restudied a passage did slightly better on a test five minutes later (81% against 75%), but students who practised recalling it did far better a week later (56% against 42%). The students who only restudied were also the most confident.
Recall even beats more elaborate study methods. In a 2011 study in Science, practising free recall beat building concept maps on a test a week later, 67% against 45%.
After a video, write down everything you remember, answer the questions from your notes, and only then rewatch the parts you got wrong.
Space it out
A review of 317 experiments found that spreading study over time beat back-to-back repetition: 47% recall against 37%. The best gap depends on when you need the knowledge: for a test a week away, a review a day or two later works well, and the gap should grow as the test gets further away, according to a large 2008 study.
Spacing also fools people. In a set of flashcard experiments, spacing beat massed study for 90% of participants, yet after the first session 72% believed massing had worked better.
A simple plan: quiz yourself on each video again the next day and about a week later. Flashcard apps with spaced repetition, such as Anki, can schedule this for you.
Mix related problems
If you are learning to solve problems, such as maths, physics or programming exercises, mix problems from different videos instead of practising one type at a time. In a four-month trial in 54 maths classes, students who practised mixed problems scored 61% on a surprise test a month later, against 38% for students who practised one type at a time. This works for problems and categories, not for memorising word lists.
Common problems, and what to do about them
"I watched hours of tutorials and still can't do it myself." This is often called tutorial hell. Stop adding videos. Take the last one you watched and rebuild what it showed without the video open. Return to it only when you are stuck.
"I keep ending up watching something else." Turn off autoplay, open videos from your playlist, and avoid Shorts while studying. If you need the home page, you will find a distraction there.
"I forget everything a week later." That is normal without review. Schedule the next-day and next-week quizzes from the section above before you close the video.
"I can't tell if the video is right." Check who made it and whether they teach or work in the field, compare a key claim with your textbook or a second source, and read the top comments for corrections.
"The video is two hours long." Use the chapters to split it into sessions. Treat each chapter as its own lesson with its own stops and questions.
A 45-minute study session from one video
- Minute 0 to 3: write the goal, skim the chapters, decide where to stop.
- Minute 3 to 30: watch at 1x to 1.5x. At each planned stop, look away, answer two or three questions, then write a short explanation in your own words.
- Minute 30 to 40: close the video. Write down everything you remember, then answer your questions without notes.
- Minute 40 to 45: rewatch only the parts you got wrong. Write the questions you want to answer tomorrow.
- Tomorrow and next week: answer those questions again before watching anything new.
Tools that help
We make Pausly, so weigh this section with that in mind.
- YouTube itself. Chapters help you plan stops, and the transcript helps you check a detail. Neither pauses the video to quiz you.
- Anki is free on computers, the web and Android (the iPhone app is paid), and the strongest option for spaced repetition if you are willing to write your own cards.
- AI summaries, from NotebookLM or similar tools, are useful for finding what a long video covers. Be careful: a summary does the thinking for you, which is the part that makes you learn. Use AI to write questions, not to replace answering them.
- Pausly automates the stops. In Focus Mode it pauses a YouTube lesson at moments chosen from the content and asks a question, then shows you where to rewatch if you miss it. You can also make a quiz from any captioned YouTube lesson or study notes from its transcript.
If you are comparing all-in-one study tools, our StudyFetch alternatives comparison covers nine of them.
Frequently asked questions
Can you actually learn from YouTube? Yes. A review of 105 studies found that adding video to existing teaching improved learning substantially. The bigger factor is whether you test yourself while and after you watch.
Should I watch YouTube lectures at 2x speed? For new or difficult material, stay at or below 1.5x. A study with real YouTube lectures found little cost at 1.5x and 2x and a clear drop at 2.5x, but a 2025 meta-analysis found that faster playback lowers retention on average.
Is a YouTube transcript enough for studying? No. A transcript tells you what was said. You learn by producing the ideas yourself: explaining them in your own words and answering questions from memory.
Should I handwrite or type my notes from videos? Either. A 2021 replication found no difference in quiz scores. What matters is turning the notes into questions and reviewing them.
How do I remember what I learned from a video? Stop every 5 to 10 minutes to answer questions from memory, test yourself again after the video, and review the next day and about a week later.
How often should I pause a YouTube lecture? Every 5 to 10 minutes, or at each chapter break. Decide the stops before you start, because people rarely pause on their own once they are absorbed.