Spaced Repetition: How to Never Forget a Word

If you’ve ever learned a word, felt confident, and then drawn a complete blank a week later, you’ve met the forgetting curve. Spaced repetition is the cure. It’s a learning method that schedules each review at the precise moment you’re about to forget, so the memory gets reinforced exactly when it needs it. The result: more words remembered, in less total study time.

How does spaced repetition work?

The core idea is simple. Memories fade on a predictable schedule, and every time you successfully recall something just before it slips away, the memory gets stronger and fades more slowly the next time. Spaced repetition turns that into a system:

• You see a word and try to recall its meaning.
• You tell the app how hard that recall was.
• The app pushes the next review further into the future each time you succeed.
• Words you struggle with come back sooner; words you know well come back rarely.

Instead of reviewing everything every day (exhausting and wasteful), you only review what’s about to fade. Over weeks, the gaps between reviews stretch from days to weeks to months, and the word effectively moves into long-term memory.

The reason this is so much more efficient than traditional study is that most of what you’d review in a “study everything” approach is wasted effort. You already know those words solidly, and reviewing them again teaches you almost nothing. Spaced repetition spends your attention where it actually pays off: on the small set of words sitting right at the edge of forgetting on any given day. A learner with 2,000 cards in their collection might only have 40 or 50 of them due for review, which is the difference between a ten-minute habit and a session you’ll quietly abandon.

The forgetting curve: why we lose words

In the 1880s, German psychologist Hermann Ebbinghaus ran experiments on himself, memorizing nonsense syllables and tracking how quickly he lost them. He discovered the forgetting curve: memory decays sharply at first, then levels off. Without review, you can lose a large share of new information within days.

Here’s a simplified picture of how retention drops for a single unreviewed item:

Time since learning	Approximate retention (no review)
Right after studying	100%
20 minutes later	~60%
1 day later	~40%
6 days later	~25%
31 days later	~20%

The exact numbers vary by person and material, but the shape is universal: the steepest losses happen early. Ebbinghaus also found the antidote. Each time you relearn the item, the curve flattens, and the forgetting slows down. That flattening is what spaced repetition is engineered to exploit.

The spacing effect: why gaps beat cramming

The spacing effect is one of the most reliable findings in learning research. Information reviewed across spaced sessions is remembered far better than the same information crammed into one block, even when total study time is identical.

Cramming feels productive because everything is fresh and fluent in the moment. But fluency in the moment is a poor predictor of memory next week. When you space your reviews, each session forces you to retrieve the word from a slightly decayed state, and that effort is exactly what builds durable memory.

Researchers sometimes call this “desirable difficulty.” A review that comes too soon is easy but barely strengthens anything, because the memory was never under any strain. A review that comes too late means you’ve already forgotten and have to relearn from scratch. The sweet spot is right before forgetting, where retrieval is effortful but still succeeds. Hitting that window by hand is nearly impossible, which is exactly why an algorithm does it for you.

A few practical takeaways:

• Short and frequent beats long and rare. Ten minutes a day outperforms an hour once a week.
• Difficulty is good. If recall feels a little hard, you’re learning. If it’s effortless, the review was probably overdue or unnecessary.
• Spreading reviews out is free retention. You don’t study more; you study at smarter moments.

Active recall: the engine behind the method

Spaced repetition only works because of active recall, the act of pulling an answer from memory rather than rereading it. Recognizing a word in a list or rereading a definition feels easy, but it does little to strengthen memory. Generating the answer yourself does.

This is why a good vocabulary card shows the word first and asks you to produce the meaning before revealing it. The moment of effortful retrieval, followed by feedback, is where the learning happens. Spaced repetition simply decides when to trigger each act of recall for maximum effect.

To get the most from active recall:

1. Look at the prompt and genuinely try to answer before flipping the card.
2. Say the meaning, or use the word in a sentence in your head.
3. Only then check the answer and rate how it went honestly.

If you want a broader playbook that combines recall with context and usage, see our guide on how to build vocabulary fast.

SM-2 vs FSRS: how the scheduling algorithm decides

The first widely used spaced repetition algorithm was SM-2, developed by Piotr Wozniak in the late 1980s. It’s elegant and still powers many flashcard tools. After each review, SM-2 multiplies your current interval by an “ease factor” that goes up when a card is easy and down when it’s hard. Simple, but it makes rigid assumptions about how everyone forgets.

FSRS (Free Spaced Repetition Scheduler) is the modern successor. Instead of a fixed multiplier, it models your memory with three components, stability (how long a memory lasts), retrievability (your current probability of recalling the card), and difficulty, and predicts when your chance of remembering will drop to a target threshold. It then schedules the review for that exact moment.

	SM-2	FSRS
Era	Late 1980s	Modern (2020s)
Approach	Fixed ease multiplier	Predicts memory state
What it estimates	Next interval	Probability you’ll recall the card
Adapts to your data	Limited	Yes, tuned to recall patterns
Typical result	More reviews	Fewer reviews, same retention

In practice, FSRS tends to ask for fewer total reviews while hitting the same retention target, because it isn’t padding intervals with guesswork. It schedules each card at the point where reviewing actually moves the needle.

It’s worth being clear about what the algorithm can and can’t do. It can’t make you remember a word you never genuinely engaged with, and it can’t read your mind, it works from the ratings you give it. But given honest feedback, it does something no human scheduler can: it tracks thousands of individual cards, each with its own decay rate, and quietly resurfaces every one of them at close to its ideal moment. You get the benefit of meticulous review planning without ever planning anything.

How Vocaby uses FSRS to time every card

Vocaby is built around FSRS so you never have to think about scheduling. The mechanics are deliberately simple on the surface:

• You study a card, recall the meaning, and then rate it with one tap.
• The four ratings are Again, Hard, Good, and Easy.
• FSRS reads that rating, updates its estimate of your memory state, and sets the next review for just before you’d forget.

Here’s what each rating signals to the scheduler:

Rating	What it means	Effect on the next review
Again	You couldn’t recall it	Comes back very soon
Hard	You got it, but barely	Comes back sooner than usual
Good	Solid, normal recall	Standard growing interval
Easy	Instant and effortless	Pushed further out

Because the timing is handled for you, a daily session is mostly just clearing the cards that are due that day. Words you’ve mastered fade into long, quiet intervals; words that are slipping resurface automatically. There’s also a fast on-demand review mode for when you simply want extra reps before a test or trip, separate from your scheduled queue.

Every card in Vocaby carries what active recall needs to work: an IPA transcription, native audio, a plain-English definition, example sentences, and synonyms. So when a word resurfaces, you’re not just memorizing a label, you’re rebuilding the full context.

Getting started with spaced repetition

You don’t need to manage any of the math to benefit from it. A simple routine is enough:

1. Pick a manageable daily volume. A handful of new words a day compounds fast over a month.
2. Do your reviews first. Clear what’s due before adding new cards, so nothing slips past the forgetting curve.
3. Rate honestly. The algorithm is only as accurate as your feedback. Don’t reward yourself with Easy when recall was a struggle.
4. Show up consistently. Streaks exist for a reason: spaced repetition rewards regularity far more than intensity.

With Vocaby you can swipe through curated topic decks, draw from a library of 29,000+ words and 167,000+ examples, get a word of the day on your home-screen widget, and add any word you meet in the wild with dictionary auto-fill. You can also browse the full collection on the Vocaby words index to see what’s waiting.

The bottom line

Spaced repetition isn’t a productivity hack; it’s how human memory actually works, organized into a schedule. The forgetting curve says you’ll lose new words fast. The spacing effect and active recall say you can keep them with surprisingly little effort, if you review at the right moments. FSRS is the algorithm that finds those moments for you, and Vocaby puts it one swipe away.

Stop relearning the same words. Start remembering them for good.

Download Vocaby on the App Store