10–20 new flashcards per day is the sustainable rate for most learners — not because new cards are hard, but because each one silently generates roughly 7–10 review sessions over the following months. Twenty new cards a day sounds modest; in six weeks it’s a ~150–200-card daily review queue. The number to manage isn’t intake — it’s the load intake creates.

The review math behind the number

A spaced-repetition scheduler resurfaces each card on an expanding ladder — in FlashDeck’s SM-2 variant, roughly day 1 → day 6 → day 15 → day 39 and multiplying from there (full table in how the algorithm works). Add a card today and it will interrupt you again at least four or five times in the next two months — more if you grade it Hard or lapse it back to the start.

That produces the steady-state ratio experienced users discover the hard way:

New cards per day	Mature daily review load	Time at ~7 sec/card
5	~40–60 reviews	5–7 min
10	~80–110 reviews	10–13 min
20	~150–200 reviews	18–23 min
30	~230–300 reviews	27–35 min
50	~380–500 reviews	45–60 min

The multiplier is honest rather than precise — lapses, Hard grades and deck difficulty push it up; Easy-heavy grading pulls it down. But the shape of the rule holds: your daily workload three weeks from now is set by your new-card rate today.

Choosing your number

• Default (languages, long-term learning): 10/day. Small enough to survive busy weeks, large enough to accumulate ~3,500 mature cards a year.
• Ambitious (full-time students): 20/day. Requires the ~20-minute daily session indefinitely — including weekends, because the scheduler doesn’t take days off.
• Deadline mode (exam in N weeks): material size ÷ weeks of runway, front-loaded. Adding 500 cards across the first three weeks then tapering beats adding 60/day for eight weeks straight — the tail of the queue needs time to mature before the test.
• Recovery mode (queue already drowning): 0/day. A week of review-only clears the backlog; the deck forgives pause far faster than it forgives overflow.

The trap to avoid is judging the habit by the new-card number. “I only added 30 cards” is how queues reach 600 due. FlashDeck enforces a per-deck daily new-card cap for exactly this reason — set it to your sustainable number in the deck settings and let the cap be the willpower. For how to structure the session itself — reviews first, honest grading, lapse handling — see study sessions.

Frequently asked questions

How many new flashcards should I add per day?

10–20 new cards per day is the sustainable band for most people. Each new card returns as roughly 7–10 scheduled reviews while it matures, so 20 new/day settles into ~150–200 reviews/day — a 20–25 minute session. Start at 10/day for two weeks, watch your due queue, then raise only if you're clearing it comfortably.

How many total cards should I review in a day?

Whatever the scheduler assigns — typically 7–10× your daily new-card rate once the deck is mature. The number to watch is time, not count: at ~5–8 seconds per card, 100 cards is about 10–13 minutes. If reviews consistently exceed 30 minutes, lower the new-card intake rather than skipping reviews — skipped dues just pile up.

Is it bad to add 50+ new cards in one day?

Not bad, but borrowed against the future. Those 50 cards re-enter the queue at 1 day, 6 days, ~15 days, ~39 days — so a binge today creates five waves of work later. It's the right move before a deadline with a planned taper; as a daily habit it compounds into a review load most people abandon. FlashDeck's per-deck new-card cap exists exactly for this.

Should I do reviews or new cards first?

Reviews first, always. Due cards are a promise the schedule already made — skipping them lengthens their overdue time and degrades the memory you paid for. New cards are optional capacity; an unlearned card today is simply a new card tomorrow. FlashDeck deals due cards ahead of new ones (interleaved ~1 new per 3 reviews) for this reason. See the session guide for the full structure.

What if I can't finish my review queue every day?

Cut intake before cutting reviews. Set new cards to 0 for a week and grind the queue down — the deck doesn't grow, but every card you do review is recovering memory that's actively decaying. A permanently overflowing queue isn't a discipline problem; it's a rate problem. The fix lives on the intake side.

Does the number change for language vocab vs exam prep?

The math is the same; the deadline changes the strategy. Open-ended learning (languages, long-term knowledge): stay at 10–15/day indefinitely. Deadline learning (a licensing exam in 8 weeks): work backwards — material size ÷ weeks, front-load new cards early so the last two weeks are pure review. Either way the ~7–10× review multiplier governs whether the plan survives contact with week three.