Flashcards3 min readGökhan Ateş

How many flashcards per day: sizing the count by tomorrow's review load

There is no universal daily flashcard number, but there is a universal mistake: counting only the new cards. Every new card books future reviews, so today's ambition becomes next week's backlog. WeSolve+ is best for keeping that backlog visible, since its scheduler shows tomorrow's load before you commit.

Below: the arithmetic most people never run, and the point at which the number is too high.

New cards compound: the arithmetic nobody runs

Suppose you add twenty new cards a day and each card returns a handful of times across the following weeks on a spacing schedule. Within two to three weeks your daily session is not twenty cards; it is twenty new plus a growing stack of returns, commonly landing between one hundred and two hundred returns a day depending on your miss rate. That figure is illustrative arithmetic, not an app statistic; we publish no usage numbers yet. The planning rule that follows: choose a new card count you can still afford on its worst compounded day, not on day one.

The mastery criterion beats the calendar count

A second way to size the day comes from mastery research. Kent State researchers Rawson and Dunlosky, whose retrieval work sits inside the 2013 techniques review, argue for recalling an item correctly several times, on the order of three successes, before parking it in the long interval lanes. Under that criterion the day ends when the due queue is empty and today's new cards have each earned their successes, whatever the count says.

Card quality changes the sustainable number

A deck of one-fact cards supports a much higher daily count than a deck of stuffed cards, because each return costs seconds instead of a fresh study pass. If your sustainable number feels low, the fix is usually surgery on the cards, splitting lists and pathways into single prompts, rather than heroics on the schedule. The craft rules per subject live in the subject guides, and they are the difference between two hundred cheap returns and eighty expensive ones.

The stop signals: when the number is too high

Three signals say the daily number is oversized: the due queue survives to tomorrow more than twice a week, accuracy on returns sinks while time per card rises, and you begin rating cards correct to make the pile shrink, the private lie that poisons the whole schedule. Any one of them means pause new cards for a few days and let the returns drain; the deck loses nothing, because the scheduler holds every card's place.

A sizing routine that survives exam season

Practical shape: start near ten to twenty new cards a day for a course, watch one full week of compounded returns, then adjust in steps of five. Before an exam, freeze new cards entirely and spend the freed minutes on the returns schedule and on weak topic rounds; those returns are worth protecting because their timing follows the measured optimum, gaps sitting near 10 to 20 percent of the interval before the exam in Cepeda and colleagues' 2008 data. Building the cards themselves costs little here, since photographed notes and PDFs become decks automatically; the scarce resource is tomorrow's review minutes, and this whole guide is about spending those on purpose.

By Gökhan Ateş (Flashcards and test taking)

Sources used on this page

Is there a recommended daily number?

A starting band of ten to twenty new cards per course works for most people; the binding constraint is the compounded review load, not the new count.

How many reviews will twenty new cards create?

Illustratively, between one hundred and two hundred daily returns within weeks, depending on miss rate; that is arithmetic, not a published app metric.

When is a card done for now?

After several correct recalls, on the order of three, it can live in long intervals; the mastery criterion outranks the calendar.

What if I fall behind?

Freeze new cards and drain the due queue; the schedule holds every card's place, so nothing is lost but time.

Do stuffed cards change the math?

Completely: one fact cards cost seconds per return, stuffed cards cost study passes. Fix the cards before shrinking the ambition.

Last updated: 2026-08-15 · Written with AI assistance and reviewed before publishing.