How to Remember Everything You Read: What Actually Works

You cannot remember everything you read, and no technique changes that. You can keep far more of what matters by changing what you do after reading: close the book, write down what you recall, check it against the text, and repeat that recall on a spaced schedule. In a 2008 experiment published in Science, learners who kept recalling material after first getting it right remembered about 80% a week later. Learners who stopped after one correct recall kept 35%.

Key takeaways

• In a 2009 survey of 177 undergraduates published in Memory, 83.6% said they reread their notes or textbook, while only 10.7% said they tested themselves.
• A 2014 meta-analysis in Psychological Bulletin found that testing beat restudying by an average effect size of 0.50, with larger gains when learners checked their answers afterwards.
• A 2008 study of more than 1,350 people in Psychological Science found the best gap before a review was about 20% of the period you need to remember something for a few weeks, falling to 5–10% for a year.
• The average adult reads English non-fiction silently at 238 words per minute, according to a 2019 meta-analysis of 190 studies in the Journal of Memory and Language.
• By the Lifoholic Recall Budget, keeping the core of a 70,000-word book costs about 56 extra minutes, roughly 19% on top of 4.9 hours of reading.

Why you forget most of what you read

Forgetting starts within minutes of finishing a page and slows over the following days. Hermann Ebbinghaus first measured this pattern in 1885, and a 2015 replication by Jaap Murre and Joeri Dros of the University of Amsterdam, published in PLOS ONE, reproduced his forgetting curve closely. Their single participant spent 70 hours learning lists and relearning them after intervals ranging from 20 minutes to 31 days.

The Ebbinghaus forgetting curve comes with a caveat that most reading advice leaves out. Both the 1885 original and the 2015 replication used nonsense syllables, which have no meaning to attach to. A chapter on interest rates connects to things you already know, so the percentages from those experiments do not measure how fast you forget a book. The shape of the curve holds. The exact numbers do not transfer.

The bigger problem is that reading feels like learning even when little is being stored. Psychologists Jeffrey Karpicke, Andrew Butler and Henry Roediger call this an illusion of competence in their 2009 study in the journal Memory: text in front of you is easy to process, and that ease is mistaken for knowing it. Every rereading makes the text more fluent, which strengthens the illusion without adding much memory.

Reading time is scarce, which makes forgetting expensive. The U.S. Bureau of Labor Statistics American Time Use Survey for 2024 found that Americans aged 75 and over read for personal interest for 46 minutes on an average day, while 15-to-19-year-olds read for 9 minutes. An analysis of the same survey by the American Academy of Arts and Sciences found that 84% of Americans read for five minutes or less a day in 2023.

Why rereading and highlighting feel like studying but aren’t

Rereading is the most popular way to study and one of the least effective. When Karpicke, Butler and Roediger asked 177 undergraduates at Washington University in St. Louis to list their study strategies in 2009, 83.6% listed rereading and 54.8% named it their number one strategy. Only 10.7% listed testing themselves, and 1.1% ranked it first.

Source: Karpicke, Butler and Roediger, Memory, 2009 (n = 177 undergraduates). Right-hand column is a Lifoholic summary of the research cited in this article.

People cannot feel the difference between the two methods while they use them. In a 2008 vocabulary experiment by Karpicke and Roediger, [LINK NEEDED: Science — Karpicke and Roediger (2008), “The critical importance of retrieval for learning”, vol. 319, pp. 966–968] learners who repeatedly recalled word pairs remembered about 80% a week later and those who stopped early remembered 35%. Every group predicted it would recall about 50%.

How to remember what you read: the close-the-book method

Retrieval practice is the act of pulling information out of memory without looking at the source. It is the most consistently supported way to make reading stick, and it takes minutes rather than hours. Christopher Rowland’s 2014 meta-analysis in Psychological Bulletin found an average effect size of 0.50 for testing over restudying, which psychologists treat as a medium effect.

Two details from Rowland’s analysis change how you should practise recall. Recall tests, where you produce the answer from nothing, gave larger benefits than recognition tests such as multiple choice. Testing with feedback produced an effect of 0.73, against 0.39 without it. Writing from a blank page and then checking the text beats both rereading and quizzing yourself with prompts.

Estimates of the size of the testing effect vary with the setting. A 2017 meta-analysis of classroom studies by Schwieren, Barenberg and Dutke found an average effect of 0.23, less than half of Rowland’s laboratory figure, largely because classrooms control less of what is tested, when and how. A 2017 review in Review of Educational Research by Adesope, Trevisan and Sundararajan, covering 272 effect sizes from 118 articles, still found practice tests more beneficial than restudying and every other comparison condition.

Recall also holds up against more elaborate study techniques. Karpicke and Janell Blunt reported in Science in 2011 that students who practised retrieval outperformed students who built concept maps, including on inference questions and on a final test that asked them to draw a concept map. A group of science educators led by J. J. Mintzes challenged the study in Science later that year, arguing that meaningful learning needs retrieval and elaboration together. The fair conclusion is that recall should come first, not that connecting ideas is useless.

The close-the-book method takes four steps:

1. Before you start a chapter, write two or three questions you want it to answer.
2. When you finish the chapter, close the book and spend three minutes writing everything you remember on a blank page.
3. Reopen the chapter and correct your page in a different colour, adding what you missed.
4. The next day, repeat step 2 from memory before looking at yesterday’s page.

Step 4 is the one most readers skip, and it is the first spaced recall. If the material builds concept by concept, the order you recall it in matters as well; Lifoholic’s explainer on how sequential learning works covers how to handle material that stacks.

How often to review a book so it sticks

The best time to review is later than feels comfortable, and the right gap depends on how long you need to remember. That is the central finding of a 2008 study in Psychological Science by Cepeda, Vul, Rohrer, Wixted and Pashler, which taught facts to more than 1,350 people, reviewed them after gaps of up to 3.5 months, and tested them up to a year later.

The spacing effect is the finding that material reviewed after a gap is remembered longer than material reviewed immediately. Cepeda’s team found that widening the gap first improved final test scores and then gradually reduced them. Measured as a share of the retention period, the best gap was about 20–40% for a one-week test and about 5–10% for a one-year test.

Source: gap ratios from Cepeda et al., Psychological Science, 2008. Day counts are Lifoholic calculations from those ratios; the one-month row applies the study’s estimate for delays of a few weeks.

Because scores declined only gradually past the best gap, reviewing a little late costs less than reviewing too early. When you are unsure which day to pick, choose the later end of the range.

The Lifoholic Recall Budget: what it costs to remember a book

The Lifoholic Recall Budget is a way to price, in minutes, what it takes to keep a book rather than just finish it. It has two formulas:

• Reading time (minutes) = word count ÷ 238
• Recall time (minutes) = (chapters × 3) + 10 for the next-day recall + 10 for each spaced review

The 238 comes from Marc Brysbaert’s 2019 meta-analysis in the Journal of Memory and Language, which pooled 190 studies and 18,573 participants and put average silent reading speed for English non-fiction at 238 words per minute. The three-minute and ten-minute recall slots are Lifoholic’s working figures, set short enough to actually do. They are not research-derived.

Here is the Recall Budget for a 70,000-word non-fiction book with 12 chapters that you need for about a month.

Source: Lifoholic calculation. Reading time uses 238 words per minute (Brysbaert, 2019); review timing uses the one-month gap ratio from Cepeda et al. (2008).

The recall routine adds 56 minutes to 294 minutes of reading, a 19% time premium. A 4.9-hour read becomes a 5.8-hour investment you can draw on later. The Recall Budget turns a vague intention to remember into a number you can schedule.

The Recall Budget also tells you where not to spend. Most books do not need one. Decide at the start whether a book is worth keeping, and reserve the 19% for those. Protected, uninterrupted blocks make the chapter recalls easier to hold to, which is the case Lifoholic makes in the deep work method for focused time, and the 10-minute next-day recall fits neatly into a consistent morning routine.

Does speed reading help you remember more?

Speed reading does not help you remember more, and it usually costs comprehension. A 2016 review in Psychological Science in the Public Interest by Keith Rayner, Elizabeth Schotter and colleagues concluded that reading involves a trade-off between speed and accuracy, and that the realistic route to reading faster with good comprehension is practice and a larger vocabulary.

The Rayner review explains why apps that flash one word at a time fall short. Many of the backward eye movements readers make are corrections for a misunderstanding. Rapid serial visual presentation, the technique those apps use, removes the chance to make them, so misreadings stay uncorrected.

Normal reading speed is slower than most people assume. Brysbaert’s 2019 meta-analysis put average silent reading at 238 words per minute for non-fiction and 260 for fiction, with most adults between 175 and 300 words per minute on non-fiction. Brysbaert found these estimates lower than the figures often cited in scientific and popular writing.

Skimming still has a legitimate place. The Rayner review notes that when thorough understanding is not the goal, skimming gets a reader through text faster with moderate comprehension. Skim the books you are sampling, and read at normal pace the ones you plan to spend a Recall Budget on.

Does sleep affect how much you remember from reading?

Sleep helps you keep what you read. Björn Rasch and Jan Born’s 2013 review in Physiological Reviews summarises more than a century of research establishing that sleep benefits memory retention. It describes sleep as a state in which recently learned material is actively consolidated, rather than simply protected from interference.

Rasch and Born highlight slow-wave sleep, the deepest stage, as important for consolidating declarative memory, which is memory for facts and events and covers what you learn from reading. An evening recall session followed by a full night’s sleep is a better use of 10 minutes than a late reading session that cuts sleep short. Lifoholic’s guide to building a restorative sleep routine covers habits that protect deep sleep.

When this advice does not apply

Most research on retrieval and spacing used students and short learning materials, not adults reading whole books for work. The 2009 survey behind the rereading figures sampled undergraduates at a selective university with average SAT scores above 1,400. The principles have held across many settings, but the size of the benefit for your own reading has not been measured.

• Reading for pleasure. If you read fiction to enjoy it, a recall routine can turn it into homework. The Recall Budget is for books you need to use.
• Reference material. Tax rules, software documentation and contracts change and can be looked up. Knowing where the answer lives is often worth more than memorising it.
• Skills learned by doing. Reading about negotiation, programming or an instrument builds knowledge, not skill. The skill comes from practice on the task itself.
• Slower or faster readers. Brysbaert found lower reading rates among older adults and people reading English as a second language, so 238 words per minute will underestimate reading time for many readers. Anyone with a reading difficulty should time a few pages and use that pace instead.
• Real-world versus lab results. Classroom studies found a smaller testing effect (0.23) than laboratory studies (0.50). Plan on the smaller benefit.
• Memory concerns. A noticeable change in how well you remember things is a question for a doctor, not a study technique.

Frequently asked questions

Is it possible to remember everything you read?

No. Forgetting is a normal part of how memory works, and even well-learned material fades without use. What you can control is which parts you keep. Choosing the few ideas that matter, recalling them without looking, and reviewing them after a spaced gap keeps those ideas available for months instead of days.

What is the best way to remember what you read in a book?

Close the book at the end of each chapter, write down everything you remember, then check it against the text. Research on the testing effect, including a 2014 meta-analysis in Psychological Bulletin, shows this beats rereading. Repeat the recall the next day, and again after a spaced gap, for the ideas you want to keep.

Does taking notes help you remember what you read?

Notes help most when you write them from memory rather than copying from the page. Copying keeps the text in front of you, which makes it closer to rereading than to recall. Write your notes after closing the book, then open it to fill the gaps. That turns note-taking into retrieval practice with feedback.

How long after reading should I review?

Review after a gap of about 20% of the time you need to remember something for a few weeks, and 5–10% for a year, based on a 2008 study of more than 1,350 people in Psychological Science. For a book you need for a month, that means one review about six days after reading. Reviewing slightly late is safer than too early.

Is highlighting a waste of time?

Highlighting is not a memory technique on its own. It marks passages you might review, which helps only if you later recall them without looking. In a 2009 survey of 177 students, 6.2% listed highlighting as a study strategy and 83.6% listed rereading. Neither replaces closing the book and testing yourself.

Why do I forget books I really enjoyed?

Enjoying a book makes it feel memorable, but that feeling is not the same as storing it. Researchers call the mismatch an illusion of competence: fluent, pleasant reading feels like learning even when little is being kept. Without recalling the ideas afterwards, the details fade whether or not you liked the book.

Try it on the next chapter you finish

The next time you finish a chapter, close the book before doing anything else and set a three-minute timer. Write everything you can remember, then reopen the chapter and mark what you missed. Tomorrow, do it again without looking. If the book is worth keeping, add one review about a week later and you have run the full Recall Budget. Small learning habits like this compound over decades, which is the argument behind Lifoholic’s piece on sustainable career longevity, and you can find more in the Career archive.

Sources and methodology

Figures were checked against the sources below on 15 September 2026. Lifoholic calculations (the Recall Budget and review-day counts) are derived from sourced figures and labelled as calculations where they appear. This article contains no affiliate links.

• Karpicke, Butler and Roediger (2009), Memory 17(4): study-strategy survey of 177 undergraduates
• Rowland (2014), Psychological Bulletin 140(6): meta-analysis of testing versus restudy
• Schwieren, Barenberg and Dutke (2017), Psychology Learning & Teaching: classroom meta-analysis of the testing effect
• Adesope, Trevisan and Sundararajan (2017), Review of Educational Research 87(3): meta-analysis of practice testing
• Karpicke and Blunt (2011), Science 331: retrieval practice versus concept mapping
• Mintzes et al. (2011), Science 334, comment: critique of Karpicke and Blunt
• Cepeda et al. (2008), Psychological Science 19(11): spacing and optimal review gaps
• Brysbaert (2019), Journal of Memory and Language 109: meta-analysis of reading rate
• Rayner et al. (2016), Psychological Science in the Public Interest 17(1): review of speed reading
• Rasch and Born (2013), Physiological Reviews 93(2): review of sleep and memory
• Murre and Dros (2015), PLOS ONE 10(7): replication of the Ebbinghaus forgetting curve
• U.S. Bureau of Labor Statistics, American Time Use Survey 2024 results (released June 2025): time spent reading by age
• American Academy of Arts and Sciences, Humanities Indicators: Time Spent Reading: analysis of reading time, 2003–2023