"How many flashcards should I do a day?" is the most common question people ask once they start, and the honest answer is that the question hides two different numbers. One is how many new cards you add each day. The other is how many reviews you do each day. They are linked, and the second one is the one that decides whether you are still using flashcards in three months.
In short
- Separate new cards per day (what you add) from reviews per day (what you do).
- Reviews per day are roughly your number of active cards divided by the typical gap between their reviews.
- Because of that, every new card adds to your workload for months, not just today.
- Start with a small number of new cards, watch your daily queue for a couple of weeks, then adjust.
- The right number is the one you can sustain through your worst week, not your best day.
Why there is no single answer
You will see rules of thumb online, anywhere from a handful of new cards a day to dozens. None of them is wrong for everyone, and none is right for everyone, because the right number depends on things the rule cannot know:
- how much time you can give each day,
- how hard your material is for you,
- how well your cards are written,
- and how long you plan to keep going.
What you can know is how the numbers fit together, which lets you pick your own.
The workload equation
When you add a card, you are not signing up for one review. You are signing up for a series of reviews that starts soon and stretches out. At first the card returns after a day or so, then after several days, then weeks, then months, as it becomes stable. (The forgetting curve explains why gaps can grow.)
A useful approximation is:
reviews per day is roughly the number of active cards divided by their average gap in days.
Here is an illustration with made-up numbers, only to show the shape. Imagine 600 cards in rotation whose average gap is 30 days: that is about 20 reviews a day. If you add more cards so that you have 1,200 with the same average gap, it is about 40 a day. Nothing about the first week told you that was coming. The reviews arrive months later, as the new cards settle into their schedule.
Two things follow:
- New cards cost the most in their first weeks. Fresh cards have short gaps, so they come back often. A big batch of new cards produces a bump of reviews soon after.
- The queue keeps growing until the pace of adding matches the pace of cards leaving daily rotation. If you add steadily, the daily load eventually flattens out, but at a level set by your intake.
So the number to watch is not today's queue. It is what the queue will look like if you keep adding at this rate for a month.
A sensible way to start
- Begin with a small intake. A handful of new cards a day is enough to feel the loop. You can always raise it.
- Review everything that is due before adding new cards. Reviews protect what you already learned; new cards are optional.
- Watch the queue for two weeks. Does it stay within the time you have? Is it growing faster than you expected?
- Adjust by small steps. Change intake slightly, then wait. Because the effect lags, a change today shows up in a few weeks.
If your queue is already heavy, pausing new cards is a perfectly good response. The backlog will melt as gaps stretch.
What affects your number
Card quality
Cards that are too large or ambiguous get missed more, and missed cards come back sooner and more often. A deck of badly written cards creates more reviews per card than a deck of clear ones. Splitting cards is often the cheapest way to cut your workload. See how to write good flashcards.
The target level of retention
If a scheduler aims to keep your chance of recall high, it brings each card back sooner. That is a deliberate trade, discussed in desired retention in FSRS.
Material difficulty
Facts that connect to things you already know are easier to keep, so their cards need fewer reviews. Arbitrary facts - a long list of unrelated codes - need more. If you are learning something genuinely new and dense, plan on a smaller intake.
Your schedule
A medical student on a clinical rotation and someone studying vocabulary at lunch have different budgets. A good intake is one that survives your busy weeks. A calmer way to study dense material is about exactly that gap between ideal plans and real weeks.
What to do when you fall behind
Everyone falls behind sometimes. The question is what you do next.
- Do not try to clear it all at once. A giant session feels virtuous but can be exhausting enough to make you quit.
- Stop adding new cards until the queue is back under control.
- Work through the queue in several short sessions across the day rather than one long block.
- Trust the scheduler. A model fit to your history, like the one described in what FSRS is, accounts for how overdue each card really is, so a backlog does not mean everything is lost.
Cirello is designed with this in mind: it shows what is actually due, not your whole library, a skipped day makes the streak plant fade instead of resetting to zero, and there is no punishment screen waiting for you. If you blank on a card, it comes back in the same session.
Getting cards in without the work
Intake is easier to control when adding a card is not a chore. If you already have a list, paste it in bulk and decide how many to start reviewing; duplicate detection means pasting an updated list later will not double what you have. Then be selective: not every item in a list needs to be in rotation today.
Reviews are short on a good day
One more reassurance: a day's reviews are usually small when your intake is moderate. A few minutes a day is a realistic target for a lot of people, and the main skill is keeping it that way. If you are curious how this plays out for particular goals, the spaced repetition flashcard app page and the guide for language learning cover typical routines.
The short answer
Add fewer new cards than you think you can handle. Review everything that is due. Check your queue weekly. Raise intake only when the queue is comfortable, and drop it when it is not. The best number of flashcards per day is the one you will still be doing next month.