Quick answer: If your daily queue feels crushing, the problem is usually not effort — it's settings, card design, or scheduler mismatch. Here's how to diagnose it. This guide gives the practical context, trade-offs, and next steps so you can decide how to apply it in your own study system.
Almost every serious Anki user hits the same wall eventually: the daily queue gets heavier and heavier, review sessions stretch past an hour, and studying starts to feel like a second job. The usual conclusion is "I just need to grind harder." That's almost always wrong. Heavy Anki queues are a symptom of fixable problems. Here are the five most common causes and what to do about each.
As of August 2026 this is still the same diagnosis. Anki 23.10+ includes native FSRS, but it stays off until you flip it on in deck options — so a collection you have been reviewing for years can still be running SM-2. If you want the settings (desired retention, interval previews, leech handling) in front of you instead of buried in desktop options, that is the gap a modern FSRS client is meant to close. The five fixes below work in Anki or Neurako; they are settings and card-design changes, not willpower.
1. FSRS isn't enabled yet
Anki has included native FSRS since version 23.10 (2023), but it stays off until you enable it in deck options. Many collections — especially older ones — still run the SM-2-based scheduler. On published benchmarks, FSRS produces roughly 20–30% fewer reviews for the same retention on typical decks. If you haven't switched yet, that alone can cut your workload meaningfully.
Fix: In Anki desktop, enable FSRS in your deck options, run the optimizer on your history (assuming you have enough reviews), and let the new schedule settle for a couple of weeks.
2. Your Desired Retention Is Too High
Many learners accept the 0.9 default and never touch it. For long-term knowledge maintenance — language learning, hobby topics, anything that isn't a dated exam — 0.9 is actually on the high side. Dropping to 0.85 or 0.87 will meaningfully reduce daily reviews at the cost of occasional forgetting (which the scheduler handles gracefully when it happens).
Fix: Drop your target by 0.03–0.05 and give it three weeks. If the forgetting feels too frequent, raise it back. You'll probably find the lower target is sustainable.
3. You're Adding Too Many New Cards Per Day
New cards compound. Every card you add today becomes a review card for months or years. A new-card limit that feels fine for a week becomes a crushing review queue six months later because intervals haven't grown enough yet to space them out.
Fix: Halve your new-card limit for a month. Let the review queue drain. Then experiment with a sustainable long-term pace. For most learners, 10–20 new cards per day is a ceiling, not a floor.
4. Your Cards Are Badly Formulated
A card that consistently takes 15 seconds to recall is a card that's testing too much at once. A card with an ambiguous prompt is a card you'll rate Again more often than you should. Both inflate review counts.
Fix: Apply Woźniak's 20 rules of knowledge formulation to your worst-performing cards. Prefer atomic cloze over long front/back. Eliminate ambiguity. Delete cards you keep failing for reasons that aren't about memory — they're just broken.
5. Leeches Are Eating Your Time
A leech is a card you've failed many times in a row. Leeches consume disproportionate review time because you keep seeing them and keep forgetting. Every SRS app has some form of leech detection, but most learners don't use it.
Fix: Enable leech detection, set a threshold (default 8 lapses is usually about right), and have the app suspend or tag leeches automatically. Then review your leech pile once a week: reformulate the cards, delete ones that aren't worth rescuing, and move on.
The Meta-Fix
Behind all five of these is one bigger principle: your spaced repetition system should serve you, not the other way around. If the queue is breaking you, the answer is not more willpower. It's better settings and better cards. Users who make the switch from Anki to Neurako often report the same thing — not that the algorithm is fundamentally different (they both run FSRS), but that the modern UI surfaces interval previews, retention settings, and leech management where you can actually see them, so you fix the workload problem before it becomes a crisis.
The workload is a knob. You have your hand on it. Use it.
Sources
- FSRS vs SM-2 benchmark: fsrs-benchmark
- 20 rules of knowledge formulation: supermemo.guru
- FSRS4Anki wiki: github.com/open-spaced-repetition/fsrs4anki/wiki