Spaced repetition: what it is and how it works for English

In Brazil, most people who have heard of spaced repetition came across it while preparing for a concurso (the competitive exams for public-sector jobs) or the vestibular (the university entrance exam). Outside that world, the term is little known, even among people who have been studying English for years. That's a shame, because the idea is simple and it solves one of the most common problems for anyone past the basics: learning a word and forgetting it a few days later.

What is spaced repetition?

It's reviewing what you learned at intervals that keep growing, where each review is an attempt to remember, not a reread. You review an expression the next day, then a few days later, then a week later, and so on. If you remembered it, the next interval gets longer. If you forgot, it comes back sooner.

Two ideas hold the method up:

• Review close to the point of forgetting. Right after you learn something, you forget quickly; with each successful review, the memory lasts longer. That's why the intervals can grow. (Why we forget is explained in Why you forget the English you study.)
• Spend time where it pays off. Instead of reviewing everything all the time, you review what you get wrong more and what you already know less.

What does a review schedule look like in practice?

Say that today you learned this sentence:

I've been meaning to call you. I've wanted to call you for a while and kept not doing it (faz tempo que eu quero te ligar).

Here's how it would play out with the intervals dailyBe uses, in days: 1, 1, 3, 5, 7, 14, 28 and 54. Each right answer moves the sentence one step up that ladder; each wrong answer moves it one step down.

These intervals aren't a scientific formula (as you'll see below, research doesn't point to a magic schedule), but they show the logic: intervals grow with right answers and shrink with mistakes. And a mistake doesn't send the sentence back to square one; it just brings it back a little sooner. After a few rounds, a sentence you once reviewed every day only shows up about every two months, and the time you free up goes to new sentences.

Notice, too, how many reviews it took: eight, spread over a little more than two months. Reviewing the same sentence eight times in a row on the first day would take about the same effort and get a much worse result.

Where does the idea come from?

From 19th-century experimental psychology, by way of cardboard boxes, all the way to software.

• Ebbinghaus, 1885. The German psychologist Hermann Ebbinghaus, in the same experiments that gave us the forgetting curve, observed that 38 repetitions spread over three days had the same effect as 68 repetitions done in a single day.
• Leitner, 1972. The German science journalist Sebastian Leitner proposed, in his book So lernt man lernen ("How to learn to learn"), a system of boxes for paper flashcards. Every card starts in the first box. If you get it right, it moves to the next box, which is reviewed less often. If you get it wrong, it goes back to the first one. In an example with three boxes, the first is reviewed every day, the second every three days and the third every five.
• Software, from 1987 on. The Polish researcher Piotr Woźniak started testing a paper review schedule in 1985 and, in December 1987, wrote the first version of SuperMemo, a program that worked out on its own when each item should come back. Since then, many flashcard programs have adopted variations of the idea.

What does the research show?

That spacing works, and that the best interval depends on how long you want to remember. The main findings:

• The effect is robust. A 2006 review led by Nicholas Cepeda pooled 839 assessments from 317 experiments, across 184 articles, on distributed practice in verbal memory tasks. The core conclusion: the gap between study sessions and the time until the final test work together, and the ideal gap grows as the time you need to remember grows.
• The ideal interval is relative. In a 2008 study with more than 1,350 people, the same team taught a set of facts, ran a review after a gap of up to 3.5 months and gave a final test up to a year later. For a test one week out, the best gap was around 20% to 40% of that time. For a test one year out, around 5% to 10%.
• It holds for second languages. A 2022 meta-analysis by Su Kyung Kim and Stuart Webb pooled 48 experiments with 3,411 participants learning a second language. Spacing had a medium-to-large effect. Longer gaps were as good as short ones on immediate tests and better on tests taken later.
• It's among the best-rated techniques. In the 2013 review of study techniques by John Dunlosky and colleagues, distributed practice and practice testing were the only two techniques rated high utility.

What are the limits of spaced repetition?

There are four limits worth knowing before you adopt the method.

1. There's no magic schedule. In Kim and Webb's meta-analysis, equal intervals and expanding intervals produced statistically equivalent results. In other words, the exact shape of the progression matters less than actually spacing your reviews and keeping them up.

2. Most of the research uses simple material. Word pairs, facts, lists. It's a good model for vocabulary, but not for everything involved in speaking a language.

3. Retaining isn't the same as using fluently. A study published in 2026 with 69 Japanese learners of English compared massed and spaced repetition for learning made-up words. Spacing improved explicit knowledge (knowing the meaning when you stop to think), but neither format developed implicit knowledge (automatic, effortless recognition). It's a single study, but the practical lesson is sensible: reviewing helps you not forget, and understanding without thinking probably also takes meeting the words in real English, on shows, in conversations and in reading.

4. The method doesn't choose what to study. Perfectly spacing a list of words you already know, or will never use, is wasted time. What goes into your reviews matters as much as the schedule. That's the topic of The words your English is missing.

How does this apply to English vocabulary?

By reviewing sentences instead of single words, and by varying the way you recall them. Compare:

The sentence shows the context, the tone and the words that usually go together. And it can be practiced in several ways: read it and understand it, hear it and understand it, see it in Portuguese and say it in English.

That's how dailyBe applies the method. The sentences you add to your list come back with spaced repetition, using the intervals in the schedule above: get one right and it moves up a step and comes back later; get it wrong and it moves down a step and comes back sooner. Reviews happen in sessions of 10 sentences, and you can do as many sessions as you like while there are sentences to study, in different modes: speaking, seeing and listening. Each word appears in several sentences, so you meet it in different contexts, and once every word in a sentence has been learned, that sentence leaves active review. The details are in How it works.

Where to start?

Pick a few sentences you actually came across and didn't understand, not a ready-made list of "useful words." Review them the next day, without looking at the answer before you try. From there, let your right answers space out the reviews and your mistakes bring sentences back. Within a few weeks, you'll notice the expressions you've reviewed starting to pop up on shows, and you'll recognize them instantly. That moment tells you whether the method is working.