Flashcards
How Many New Flashcards Per Day? A Realistic Limit
How many new flashcards per day should you add? A simple review-load table, the research behind spacing, and a rule for picking 5, 10 or 20 new cards.
By iranimij Published 7 min read
For most learners the best answer to “how many new flashcards per day?” is 10: enough to build a vocabulary of about 300 words a month, small enough that the reviews stay manageable. Use 5 on busy weeks and 20 only if you have the time to review everything those cards create.
The number that really decides your day is not how many cards you add, but how many reviews come back. This post shows how to estimate that pile, what the research says about spaced flashcards, and how to choose your own limit.
How many new flashcards per day is right for you?
Start at 10 new cards a day, review first, then add new ones. Then adjust. If you clear every review in the time you have, raise the number by 5. If the review pile keeps growing for a week, lower it by 5.
Here is a quick guide to the three common settings:
| New cards per day | New words per month | Best for |
|---|---|---|
| 5 | about 150 | Busy weeks, a second language alongside work, getting back into the habit |
| 10 | about 300 | The default for adult learners with 15–20 minutes a day |
| 20 | about 600 | Intensive study with 30+ minutes a day and a plan for the reviews |
These are planning figures, not research results. No study has found one universal best number of new cards, because the right limit depends on your time, your cards and your language.
Why do the reviews matter more than the new cards?
Every new card is a promise to review it several more times. Spaced repetition, which means reviewing material at growing intervals, shows you each card again just before you would forget it. One new card today therefore becomes a handful of reviews spread over the next weeks and months.
Anki’s manual puts a number on this: if you are consistently learning 20 new cards a day, you can expect about 200 reviews a day (Anki Manual). That is a rule of thumb for Anki’s own scheduler, and it varies with how often you miss cards.
You can see the effect yourself with a simple model. Take a seven-box Leitner schedule with review gaps of 1, 2, 4, 8, 16, 31 and 120 days, and assume you answer every card correctly. A card is then reviewed on days 1, 3, 7, 15, 31, 62 and 182: seven reviews in about six months. That makes the daily load roughly seven times your new-card number once the system has filled up.
| New cards per day | Reviews per day after about 6 months (simplified model) | Time at ~8 seconds per review |
|---|---|---|
| 5 | about 35 | about 5 minutes |
| 10 | about 70 | about 9 minutes |
| 20 | about 140 | about 19 minutes |
| 40 | about 280 | about 37 minutes |
This is my own arithmetic on a simplified model, not a measurement. Real decks have misses, which send a card back toward the start and add reviews, and the 8-second figure is an assumption you should replace with your own pace. The shape of the result is what matters: doubling your daily new cards doubles your daily reviews, and the bill arrives a few weeks later, when the early enthusiasm has faded.
What does the research say about spaced flashcards?
Research supports reviewing flashcards on a spread-out schedule and testing yourself, but it does not name a perfect number of cards. Two findings explain why the daily habit matters more than the daily quota.
Nate Kornell asked participants to learn word pairs either from one large stack, which spreads each card’s repetitions further apart, or from four small stacks studied one after another. Spacing was more effective for 90% of participants, yet after the first session 72% believed that cramming had worked better (Kornell, 2009). Spacing feels harder, so you cannot judge your own deck by feel. Our post on spaced repetition vs cramming goes through the wider evidence.
Karpicke and Roediger had students learn 40 Swahili-English word pairs. Continued testing produced a large benefit for delayed recall, and students’ predictions of their own performance were uncorrelated with their actual results (Karpicke & Roediger, 2008). A later study questioned how little re-studying helps, so treat the exact size of the effect with care, but retrieval practice remains one of the best-supported techniques.
Put together, these results point one way: what helps is showing up for the reviews. A limit that lets you finish every review beats a larger one you abandon.
How do you choose your own limit?
Work backwards from the minutes you have, not forwards from the words you want. Follow this method:
- Decide how many minutes a day you can protect, even on a bad day. Say 15.
- Time yourself on 20 reviews to find your seconds per card.
- Divide your minutes by your seconds to get the review load you can carry: 15 minutes at 8 seconds is about 110 reviews.
- Divide that by 7, the rough multiplier from the table above. You get about 15 new cards a day at the extreme, so choose 10 to leave a margin for misses and busy days.
The margin matters. Plan for the day you are ill, travelling or tired, because that is the day the pile grows.
Your cards also change the answer. Piotr Wozniak, creator of SuperMemo, describes the minimum information principle: keep each card as simple as possible, with a short answer (Wozniak, 1999). A card with one word and one short example takes seconds to review; a card with a paragraph takes minutes. Simpler cards let you carry more of them.
Where does Radicard fit in?
Radicard keeps the daily number small by design. You photograph any text, such as a café menu, a street sign or a book page, and Google’s Gemini AI picks out 5–15 vocabulary words matched to your CEFR level (A1–C2), with translations. CEFR is the European scale that rates language skill from A1 to C2. A single scan lands you right in the 5–15 range recommended above.
The words are saved on your device and reviewed with the Leitner system, using seven boxes with gaps of 1, 2, 4, 8, 16, 31 and 120 days. You do not set a limit or tune a scheduler. Reviews work offline; scanning a new photo needs internet, and the free plan includes 5 scans a day. Radicard builds your deck from photos you take, not from a pre-made course. See how it works or try the web app in any browser; it is also on the App Store and Google Play.
What should you do when your reviews pile up?
Pause new cards, keep doing the reviews, and let the pile shrink. Reviews are the part of the system that protects what you already learned. New cards can wait.
- Stop adding new cards for three to five days, or until the daily pile feels normal.
- Do reviews in short sessions, for example 10 minutes twice a day, instead of one long catch-up.
- Cut cards that do not matter. If a card is about a word you never meet, delete it.
- Do not reset your deck. Starting over throws away the intervals you already built.
If you are comparing how different schedulers handle misses and backlogs, read FSRS vs SM-2 vs Leitner. The box mechanics themselves are explained in the Leitner system explained.
Should the limit change for different kinds of words?
Yes: lower it for hard material and raise it for easy material. Hard cards cost more reviews and more time.
- Kanji, Arabic root words or long compound nouns: start at 5 to 10, because each card takes longer and fails more often. See learning kanji with spaced repetition.
- Cognates and loanwords: you can add more, since many words are already half-known.
- Words from your own photos or reading: these tend to stick better because they connect to something you saw, which is why our guide on remembering vocabulary recommends personal context.
What to do next
- Set your new-card limit to 10 a day (5 if your week is busy) and write it down.
- Always do your reviews first, then add new words.
- After two weeks, check whether you finished every review each day. If yes, raise the limit by 5. If no, lower it by 5.
- When the pile grows, pause new cards for a few days instead of skipping reviews.
- Keep each card to one word and one short example to make reviews quick. To pick an app, compare options in the best flashcard apps for language learning.
Frequently asked questions
How many new flashcards should I learn per day?
How many new cards per day is Anki's default?
What happens if I add too many new flashcards?
Is it better to learn fewer cards well or many cards quickly?
Keep reading
Flashcards
Reverse Flashcards: Should You Test Both Directions?
Should you make reverse flashcards? See what research says about testing words from the new language and from your own, and a simple rule for where to start.
6 min read
Flashcards
FSRS vs SM-2 vs Leitner: Which Flashcard Algorithm Wins?
FSRS vs SM-2 vs Leitner compared: how each flashcard scheduler decides when you review, what the evidence says, and which one fits how you study.
6 min read
Flashcards
The Leitner System Explained: Why Boxes Beat Cramming
The Leitner system explained: how a simple set of flashcard boxes spaces your reviews, how to build one in ten minutes, and where it beats cramming.
6 min read