Spaced repetition is the most evidence-backed memory technique available. The research is decades old, the effect sizes are large, and the mechanism is well-understood: reviewing material at increasing intervals — just before you would otherwise forget it — forces retrieval that strengthens memory far more than massed review or re-reading.
The problem is that most of the tooling built around spaced repetition — Anki, SuperMemo, Remnote — is designed for discrete, atomic facts. Vocabulary words, historical dates, chemical formulas. This works brilliantly for those domains.
It is less obvious how to apply spaced repetition to learning from YouTube video lectures, where the content is continuous, contextual, and often procedural rather than factual. Nobody has yet made an Anki deck for "understanding why dynamic programming works."
This post covers a practical system for applying spaced repetition principles to YouTube-based learning — one that works with the nature of video content rather than forcing it into flashcard format.
The Forgetting Curve and Why It Matters for YouTube Learners
Hermann Ebbinghaus's forgetting curve, first documented in the 1880s and replicated consistently since, shows that memory of new material decays rapidly without review. Without any review, roughly 50% of new information is forgotten within an hour, 70% within 24 hours, and more than 80% within a week.
For YouTube learners — who typically watch a lecture once and move on — this means the majority of what they watched is effectively gone within a day. The sensation of understanding during the lecture is real; the retention a week later is not.
Spaced repetition attacks this directly by scheduling reviews to happen just before the forgetting curve would otherwise take over. One review 24 hours after initial learning, a second review 3-5 days later, a third review 2 weeks later. Each review resets the forgetting curve at a higher baseline, producing retention that persists for months or years rather than days.
The challenge for YouTube learners is that the "material to review" is not a set of flashcards — it is a set of concepts encountered in a video context, with procedural and structural knowledge that does not reduce to question-answer pairs.
The System: Review Without Re-Watching
The core principle: spaced repetition for YouTube learning is implemented through note review and retrieval practice, not through re-watching lectures.
Re-watching is the common instinct and the wrong one. Re-watching a lecture is recognition practice — you see the content again and it feels familiar. Reviewing your notes and trying to reconstruct the concepts from memory is retrieval practice — you try to produce the knowledge without seeing it, which is what strengthens the memory trace.
The system has three components.
Component 1: Structured notes during the initial watch
This is the capture layer, and it determines the quality of everything that follows. In Courseifier, take notes in the integrated panel with a consistent structure: the main concept in one sentence, the mechanism or key insight, one example, and any relevant equations or code.
The crucial element: at the bottom of each lecture's notes, write three to five questions the lecture answers. These questions become your spaced repetition prompts. "What is the space complexity of merge sort and why?" "What distinguishes a B-tree from a binary search tree?" "When would you prefer a heap over a sorted array?"
These questions should require genuine recall to answer — not "what is the name of the algorithm that runs in O(n log n)" but "why does merge sort's recurrence T(n) = 2T(n/2) + O(n) resolve to O(n log n)?"
Component 2: The review schedule
After finishing a lecture, schedule three reviews:
- Review 1: the following day
- Review 2: five days after Review 1
- Review 3: two weeks after Review 2
For each review, open your notes and try to answer the questions you wrote without looking at the answers. Write your answers in a separate document or on paper. Then compare to your notes.
If you answered correctly: extend the next review interval.
If you answered partially: review the note, schedule a shorter next interval.
If you could not answer: re-read the note carefully, then attempt to re-explain the concept in your own words before scheduling a shorter next interval.
This is manual spaced repetition — it requires tracking your review schedule. A simple approach: a dated table in Notion or Obsidian, or a plain text file with lecture names and upcoming review dates. It does not need to be sophisticated.
Component 3: Export and consolidate
When you complete a course in Courseifier, export all notes as a Markdown file. Import into your PKM tool. During your periodic reviews, you now have a structured document for each course — every lecture, every question, organised chronologically.
After three or four review cycles per lecture, the material has typically been retained well enough that the question-answer format becomes redundant. The knowledge is accessible under different framings, in different contexts, when applied to problems — which is what genuine understanding looks like.
Anki Integration for Factual Content
For the subset of DSA and technical knowledge that reduces to discrete facts — time and space complexity of standard algorithms, definitions, specific theorem statements — Anki is a useful complement to the system described above.
After each lecture review, create Anki cards for any facts you repeatedly answer incorrectly in your note-based review. The card format that works best for technical content:
Front: "What is the time complexity of building a heap from an unsorted array, and why is it not O(n log n)?"
Back: "O(n) — because most elements are near the bottom of the heap where sift-down operations are cheap (O(1) for leaf nodes). The majority of work happens at low levels. Formal proof via summation of levels × nodes-at-level converges to O(n)."
These cards are not a replacement for conceptual understanding — they are a supplement that ensures the specific facts support your recall of the broader concept. The conceptual understanding lives in your Courseifier notes and exports. The specific facts live in Anki.
What Spaced Repetition Cannot Do
Spaced repetition is a retention technique. It ensures you remember what you learned. It does not generate understanding you never had in the first place.
If your initial capture of a concept was shallow — you watched the lecture but the underlying logic did not click — spaced repetition will efficiently preserve that shallow understanding. You will accurately remember that merge sort is O(n log n) without being able to derive it or explain it.
The quality of the initial encoding determines the quality of what gets retained. This is why the active recall technique described in our post on how to use active recall when learning from YouTube is a prerequisite to effective spaced repetition — the retrieval practice during the initial watch produces the depth of encoding that spaced repetition then preserves.
The full system: active recall during the lecture (pause and retrieve cycle), structured notes with questions in Courseifier, spaced review of those notes at increasing intervals, Anki for specific facts. Each component addresses a different part of the learning and retention problem.
For the study environment that makes this system sustainable — removing the distractions that interrupt both the initial encoding and the review sessions — see how to study from YouTube without distractions.