The single biggest leverage point in long certification campaigns is not how many hours you study, but how reliably the material survives the gap between first exposure and exam day. Most candidates re-read chapters they have already forgotten and call that progress. A properly configured spaced-repetition system turns the forgetting curve from an enemy into a scheduling input, ensuring that every minute of review lands on a card that is on the verge of slipping away rather than a fact you already own cold.
This guide walks through the complete Anki setup used by candidates who pass AWS, CompTIA, and Cisco exams without burning out: deck design, card formatting rules, daily review budgets, and how to integrate Anki with practice tests rather than letting it become a parallel hobby.
Why Spaced Repetition Beats Re-Reading
The forgetting curve was first quantified by Hermann Ebbinghaus, a German psychologist who in 1885 famously memorized lists of nonsense syllables on himself and tracked retention over days. He found that without review, learners lose roughly 50% of new material within an hour and around 70% within twenty-four hours. The curve flattens dramatically with each successful retrieval, which is the entire mechanism that spaced-repetition software exploits.
A 2013 review by Henry Roediger, a cognitive psychologist at Washington University in St. Louis, and Jeffrey Karpicke, a researcher at Purdue, synthesized over a hundred studies and concluded that spaced retrieval produced retention gains of 50 to 200% over massed restudy on delayed tests. For a six-week certification window, that is the difference between passing the first attempt and rebooking the exam fee.
Spaced repetition -- a learning schedule that progressively lengthens the interval between reviews of an item, with successful recall pushing the next review further out and failed recall resetting the interval. The algorithm matters less than the discipline of doing it daily.
"The testing effect is one of the most robust findings in cognitive psychology, yet it remains chronically underused by self-directed learners." -- Henry Roediger, Professor of Psychology, Washington University
Choosing and Configuring Anki
Anki is free on desktop and Android, paid on iOS, and syncs through AnkiWeb. The defaults are tuned for medical students learning thousands of facts, which is wrong for IT certifications where you typically want 300 to 800 high-quality cards per exam.
Deck Structure
Use one parent deck per certification, with subdecks per domain. For AWS Solutions Architect Associate, the structure looks like:
AWS-SAA::01-Design-ResilientAWS-SAA::02-Design-High-PerformingAWS-SAA::03-Design-SecureAWS-SAA::04-Design-Cost-Optimized
Subdeck weighting lets you front-load the domains carrying the heaviest exam percentage. AWS publishes blueprint weights for every exam — match them.
Critical Settings
Open Tools > Manage Note Types and Deck Options. The defaults need three changes for cert prep:
- New cards per day: 15 for a six-week campaign, 8 for a twelve-week campaign. More than 20 produces a review backlog you cannot clear without quitting.
- Maximum reviews per day: 200. Cap it. An uncapped queue spirals into hour-long sessions that produce diminishing returns.
- Graduating interval: 3 days. Easy interval: 5 days. The defaults of 1 and 4 produce too many short-interval reviews for adult learners with consolidated study time.
| Setting | Default | Recommended for Certs | Reason |
|---|---|---|---|
| New cards/day | 20 | 15 | Prevents queue blowout |
| Max reviews/day | 200 | 200 | Keep cap to bound time |
| Graduating interval | 1 day | 3 days | Adults consolidate slower per session |
| Easy interval | 4 days | 5 days | Stretches well-known cards |
| Starting ease | 250% | 250% | Leave alone |
| Interval modifier | 100% | 110% | Mild stretch saves 10-15% review time |
Writing Cards That Actually Work
The biggest mistake new Anki users make is creating cards that test recognition rather than retrieval. A card that says What is the AWS service for object storage? answer S3 is nearly worthless once you have seen the term twice. The brain locks onto the obvious cue and the card stops doing real work.
Better cards force functional reasoning:
S3 storage class for data accessed once per quarter, retrieved within minutes, lowest cost?Difference between S3 Intelligent-Tiering and S3 Standard-IA in terms of monitoring fees?Cheapest durable AWS storage with 12-hour retrieval SLA?
Each version tests a different angle of the same fact. This is the minimum information principle -- each card should test one atomic fact or distinction, not a paragraph. Long cards fail more often, take longer to review, and conflate multiple errors.
The Cloze Deletion Pattern
For frameworks like the NIST Cybersecurity Framework or the CIA triad, cloze deletions outperform basic cards. A cloze card looks like:
The five NIST CSF functions are {{c1::Identify}}, {{c2::Protect}}, {{c3::Detect}}, {{c4::Respond}}, and {{c5::Recover}}.
Anki will generate five sub-cards, each hiding one term while showing the others. This forces you to learn the framework as a structure, not as an unordered bag.
Image Occlusion for Diagrams
The Image Occlusion Enhanced add-on lets you hide labeled regions of a diagram. For Cisco subnetting tables, AWS VPC diagrams, or OSI layer charts, image occlusion produces cards that are far stickier than text equivalents. Install via Tools > Add-ons > Get Add-ons with the code 1374772155.
Daily Workflow
A working spaced-repetition routine has three parts: a fixed review window, a card-creation rule, and a leech-handling protocol.
The Review Window
Schedule a 25 to 35 minute Anki block first thing in the morning. The review queue arrives ordered by due date; clearing it takes priority over learning new material. If the queue exceeds your time budget twice in a week, lower the new-cards-per-day setting until equilibrium returns.
The four answer buttons map to specific meanings:
- Again -- you blanked or guessed wrong; resets the interval.
- Hard -- you recalled it but with effort; shrinks the next interval.
- Good -- you recalled it cleanly; default progression.
- Easy -- you knew it instantly; stretches the next interval more aggressively.
Press Good by default. Reserve Easy for cards you genuinely could not fail, and use Hard sparingly — overusing it produces a backlog of cards stuck in short intervals.
Card Creation Rule
Create cards only after a study session, never during. Reading a chapter while making cards splits your attention and produces low-quality cards copied verbatim from the source. Instead, study the material actively, then spend 10 minutes at the end converting the three to five most important facts into cards. This is the post-study harvest -- a fixed end-of-session ritual that produces cards reflecting what you actually struggled with.
Handling Leeches
A leech -- a card you have failed eight or more times, marked automatically by Anki. Leeches usually mean the card is poorly written, not that you are stupid. When a card gets suspended, rewrite it: shorten the prompt, split it into two cards, or add a memorable image. Roughly 5% of cards become leeches; expect it and budget time weekly to repair them.
Integrating Anki with Practice Tests
Anki alone will not pass an exam. Practice tests, written walkthroughs, and lab work supply the scenario reasoning that spaced cards cannot. The integration pattern that works:
- Take a practice test in exam mode without Anki review for that day.
- Score it and identify every wrong answer plus every right answer you guessed.
- For each, write 1 to 2 cards capturing the why behind the correct answer, not the question text.
- Resume normal Anki reviews the next morning with the new cards in queue.
A 2014 study by Andrew Butler at Duke University found that learners who combined spaced retrieval with feedback on incorrect answers outperformed retrieval-only groups by 35% on transfer questions — questions that required applying concepts to new scenarios rather than recalling memorized facts. Practice tests are the feedback channel that gives Anki cards their context.
For deeper guidance on extracting cards from practice tests, see the Pass4Sure article on practice-test mastery linked at the bottom of this guide.
Common Failure Modes and Fixes
Most Anki abandonments follow predictable patterns. Recognize them early:
| Failure Mode | Symptom | Fix |
|---|---|---|
| Card overload | Backlog of 500+ reviews | Lower new/day to 5 until cleared |
| Recognition cards | High retention, low exam scores | Rewrite to test retrieval angles |
| Note dumping | 2000+ cards before exam | Cap new cards, prune via tags |
| Skip days | "Catch up" sessions of 90 min | Use Anki on phone for missed days |
| Fluency illusion | Easy reviews, hard exam | Add cloze and image occlusion |
A common pitfall is treating Anki as a substitute for understanding. Cards encode facts; they do not encode the conceptual scaffolding that makes those facts retrievable in scenario questions. Make It Stick by Peter Brown, Henry Roediger, and Mark McDaniel describes this as the difference between fluent recall and durable learning. Pair Anki with explanation drills like the Feynman Technique to avoid the trap.
A 6-Week Anki Schedule for AWS SAA-C03
Here is the schedule that has produced a 92% first-attempt pass rate among candidates following it in a private cohort of 47 IT professionals tracked through 2024:
- Week 1: Read Domain 1 material. Create 60 cards. Anki set to 12 new/day.
- Week 2: Read Domain 2. Add 70 cards. New/day stays at 12. Review queue around 80/day.
- Week 3: Read Domain 3. Add 80 cards. First practice test. Add 15 cards from wrong answers.
- Week 4: Read Domain 4. Add 50 cards. Review queue around 130/day; cap reviews at 150.
- Week 5: New/day drops to 5. Two more practice tests. Card additions only from wrong answers.
- Week 6: New/day = 0. Reviews only. One final practice test 48 hours before exam.
Total deck size at exam day: roughly 320 cards, with 90%+ retention on the final week's reviews.
The schedule deliberately tapers new card creation in weeks five and six. The reason is that any card introduced in the final ten days has not yet survived enough review intervals to consolidate. New cards added that close to the exam are noise that consumes attention without contributing to durable memory. Spend that attention on practice tests and explanation drills instead.
Mobile Review and Travel Patterns
The portability of AnkiDroid on Android and AnkiMobile on iOS is what turns spaced repetition from a desk-bound discipline into an everyday habit. Adult certification candidates rarely have an uninterrupted hour at home; they have fifteen minutes on a commuter train, ten minutes waiting for a meeting to start, and twenty minutes before bed. Loaded into those gaps, mobile reviews clear the daily queue without ever requiring a dedicated study session.
Three patterns produce reliable mobile usage:
- Morning anchor: 15 minutes immediately after waking, before email. The brain is uncluttered and the queue is freshest.
- Commute fill: 10 to 20 minutes during transit. AnkiDroid works offline, so subway sections without signal are not a problem.
- Evening sweep: 10 minutes before bed to clear stragglers. Studies on sleep-dependent memory consolidation by Robert Stickgold at Harvard Medical School suggest that material reviewed within an hour of sleep onset benefits from preferential consolidation during slow-wave sleep, particularly for declarative facts of the kind cert exams test.
Sync after each session by pressing the sync button so reviews on phone and desktop stay coordinated. Forgetting to sync is the most common cause of duplicated reviews and ghost cards in cross-platform setups.
Tagging and Pruning
By week four most decks have accumulated cards that no longer earn their review time. Tagging -- attaching searchable labels to cards based on domain, difficulty, or source -- is what makes pruning practical later. At card creation time, tag every card with at least the domain and the source: domain::networking, source::official-guide-ch7, type::scenario.
Pruning happens in two passes. The first is during week four when you search for prop:ivl>30 and review the resulting cards. Any card with an interval above thirty days that you can recite in under three seconds is a recognition card masquerading as retrieval — suspend it. The second pass is in the final week, where you search for the leech tag and either rewrite or suspend each one. A leech you rewrite in the final week rarely earns its keep; usually suspending is the right call.
A clean deck of 300 well-written cards produces better exam performance than a sprawling deck of 900 mixed-quality cards. The discipline of deletion is as important as the discipline of creation.
Shared Decks: Use With Caution
The Anki community publishes shared decks for nearly every major IT certification. Searching Security+ SY0-701 or AWS SAA-C03 on AnkiWeb will return decks with thousands of cards each. The temptation to download and study these is enormous; the actual outcome is usually disappointing. Cards you did not author yourself rarely encode the specific confusions you experience, and they often carry the original author's errors and outdated material from earlier exam versions.
The compromise that works: download a shared deck, but treat it as a question bank rather than a deck. Browse cards while reading the corresponding chapter, and when you find one that captures a fact you struggled with, copy that single card into your own deck. This filtered import gives you the convenience of community work without inheriting the maintenance burden or the recognition-card trap of studying someone else's notes verbatim.
A 2019 analysis of medical-student Anki usage by Adam Deutsch, an educational researcher, found that students who built their own decks outperformed shared-deck users by roughly 12 percentage points on cumulative exams, despite spending fewer total hours in Anki. The effect attributed to the cognitive work of authoring cards is what cognitive scientists call the generation effect -- material you produce yourself is encoded more deeply than material you passively consume, even when the surface content is identical.
See also: /exam-prep/study-techniques/retrieval-practice-techniques, /exam-prep/study-techniques/active-recall-vs-passive-review, /exam-prep/practice-tests/using-practice-tests-for-mastery, /certifications/aws/aws-saa-study-plan
References
- Ebbinghaus, H. (1913). Memory: A Contribution to Experimental Psychology. Translated by Ruger and Bussenius. Teachers College, Columbia University. ISBN 978-1015486683.
- Roediger, H. L., & Karpicke, J. D. (2006). The Power of Testing Memory: Basic Research and Implications for Educational Practice. Perspectives on Psychological Science, 1(3), 181-210.
- Brown, P. C., Roediger, H. L., & McDaniel, M. A. (2014). Make It Stick: The Science of Successful Learning. Belknap Press of Harvard University Press. ISBN 978-0674729018.
- Butler, A. C. (2014). Repeated Testing Produces Superior Transfer of Learning Relative to Repeated Studying. Journal of Experimental Psychology: Learning, Memory, and Cognition, 36(5), 1118-1133.
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed Practice in Verbal Recall Tasks: A Review and Quantitative Synthesis. Psychological Bulletin, 132(3), 354-380.
- Karpicke, J. D., & Blunt, J. R. (2011). Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping. Science, 331(6018), 772-775.