Buy Sumizeit infographics

How to Remember What You Read

Posted on 7/1/2026, 10:59:57 AM

Most people forget 90% of what they read within a week. Here's the science of why it happens and the simple system — including Sumizeit's spaced repetition flashcards — that fixes it for good.

Share this article

For 15-minute non-fiction book summaries of best-selling books, check out sumizeit.com

TL;DR

Most people forget 90% of what they read within a week. The problem is not intelligence or attention span — it is the way reading is typically structured, which creates no mechanism for retention. The solution combines a few well-researched techniques: reading with intention, taking active notes, revisiting key ideas at spaced intervals, and using tools like Sumizeit's flashcard feature to automate the hard parts of remembering. This article explains why forgetting happens and exactly how to stop it.


The Forgetting Problem Nobody Talks About

Here is something that happens to almost every reader at some point. You finish a book you genuinely found useful. You felt engaged while you were reading it. You dog-eared pages, maybe underlined passages, possibly told someone about it. Three months later, a friend mentions the book and asks what you thought of it. You remember that you liked it. You remember approximately the topic. You cannot recall a single specific idea, framework, or takeaway with enough clarity to explain it to another person.

This is not a personal failure. It is a near-universal experience, and it has a name. Hermann Ebbinghaus, a German psychologist working in the 1880s, mapped what he called the forgetting curve: the rate at which newly learned information decays in the absence of reinforcement. His findings were sobering. Without any review, people forget roughly 50% of new information within an hour of encountering it. Within 24 hours, the number climbs to around 70%. Within a week, close to 90% of the material is effectively gone.

The implication for readers is significant. If you read a book over two weeks and then never revisit the material, you are retaining approximately 10% of what you engaged with — and the 10% that survives is rarely the most important part. It is the pieces that happened to align with something you already knew or that you've coincidentally encountered again since.

The good news is that the forgetting curve is not fixed. Ebbinghaus also identified what counteracts it: retrieval practice, distributed across time. Every time you actively recall a piece of information, the forgetting curve resets and flattens. Information that has been retrieved several times at spaced intervals becomes genuinely durable — the kind of knowledge that is available to you months or years later, in real conversations and real decisions, not just in the hours immediately after you read it.

Understanding this changes how you think about reading. The goal is not to get through books. The goal is to get ideas from books into your long-term memory, which is a different activity with a different set of requirements.


Why Reading Alone Is Not Enough

The way most people read is passive. They move through text from left to right, absorbing as they go, occasionally pausing when something is interesting or confusing, and then moving on. This feels like learning because engagement with new ideas produces a subjective sensation of understanding. But the feeling of understanding and the actual encoding of durable memory are two different things, and passive reading reliably produces the former without reliably producing the latter.

The cognitive science here is well established. Memory is strengthened by what researchers call "desirable difficulties" — conditions that make the learning slightly harder in the moment but significantly more durable over time. Passive reading is too easy. Your brain processes the text without being asked to do anything with it, which means it flags the information as not particularly important to store.

Active engagement, by contrast, creates conditions under which the brain takes the information seriously. This is why you remember the answer to a question you struggled with better than the answer to one that came easily. It is why taking notes in your own words produces better retention than highlighting the author's words. It is why testing yourself on material, even when you get the answers wrong, produces better long-term retention than re-reading the same material.

The implication is not that passive reading is worthless. It is that passive reading without any reinforcement strategy is reading that mostly doesn't stick.


The Techniques That Actually Work

Read with a question in mind. Before you open a book or a chapter, decide what you are trying to find out. This sounds trivially simple and its effect on retention is not trivial. When you have a question in your mind, your brain filters the incoming text for information relevant to that question and tags it as important. The chapters you read with a specific question retain far better than the chapters you read because you were on page 94 and it was time to read page 95.

The question does not need to be sophisticated. "What is the central argument of this chapter?" is enough. "How does this apply to the decision I'm currently trying to make?" is better. "What would I tell someone else about this?" is better still, because it primes your brain to encode the material in the form in which you'll actually use it.

Take notes in your own words. This is the single highest-leverage reading habit available to most readers, and most readers don't do it. The act of translating an author's idea into your own language requires you to actually understand it rather than just recognize it, which is a fundamentally different cognitive operation. Recognition feels like knowledge but is not the same thing. Generation — producing the idea from your own mental resources — is what builds durable memory.

Your notes do not need to be long or beautifully organized. A sentence or two per major idea, written in plain language, is enough. The point is the translation, not the archive. Even if you never read your notes again, the act of writing them has already done the work.

Talk about what you read. Explaining an idea to another person is one of the most powerful memory-encoding activities available. It forces retrieval, requires you to organize the material into a coherent explanation, and exposes gaps in your understanding that re-reading often conceals. The person you explain it to doesn't need to be particularly interested. You are not doing this for them.

The practice of reading a chapter and then explaining it aloud to yourself — or to a spouse, a friend, a pet with resigned patience — before moving on produces significantly better retention than simply reading the next chapter.

Use spaced repetition. This is where the research is clearest and where most readers' practice diverges most sharply from what actually works. Revisiting material at increasing intervals — after one day, then three days, then a week, then two weeks — produces dramatically better long-term retention than either massing all your review into one sitting or not reviewing at all. The spacing forces retrieval at the exact moments when forgetting is about to occur, which is the condition under which memory is most durably reinforced.

The challenge with spaced repetition done manually is that it requires you to track when you last reviewed each piece of information and when you should review it again. This is feasible for people who are deeply organized and enjoy maintaining review systems. For everyone else, it is a friction-generating overhead that causes the practice to quietly die within a month.

This is where Sumizeit comes in.


How Sumizeit's Flashcard Feature Changes the Equation

Sumizeit is a book summary app that delivers 15-minute summaries of bestselling nonfiction across text, audio, video, and infographic formats. But the feature that addresses the retention problem directly is its spaced repetition flashcard system.

After you complete a summary on Sumizeit, the app automatically generates a set of flashcards containing the book's key concepts, frameworks, and most important takeaways. You don't have to create the cards yourself. You don't have to decide what's worth remembering. The cards are ready immediately, built from the ideas that matter most in the summary you just read.

The cards then enter a spaced review queue powered by the SM-2 algorithm — the same algorithm that underpins Anki, the most widely used spaced repetition system in the world. The algorithm tracks your responses for each card and schedules your next review at the optimal interval. If you recalled something easily, the next review is pushed further into the future. If you struggled, the card comes back sooner. Over time, the algorithm builds a review schedule calibrated to your actual memory, not to an arbitrary interval someone guessed in advance.

The result is that the ideas from the books you read through Sumizeit don't fade the way they normally would. They get revisited at exactly the right moments — enough times to move from the short-term holding area where most reading ends up into the durable long-term memory where the ideas are actually useful.

The practical experience of this is straightforward. You read a summary during a lunch break or a commute. That evening you spend two or three minutes on the flashcard review that Sumizeit queues for you. Three days later, a few more cards appear in your daily review. A week after that, a handful more. The daily review takes less than five minutes for most users. The cumulative effect, across months of consistent use, is a genuine and measurable expansion of what you know and can recall.

This is the mechanism that the moms, dads, executives, and students who read 20 or 30 or 50 books a year are using to actually retain what they read — not just accumulate a count.


Building a Reading System That Sticks

Retention is not something that happens to you after reading. It is something you build into the act of reading itself. The readers who remember what they read have not discovered some innate memory superpower. They have installed a small set of habits that make remembering the automatic byproduct of how they read, rather than something they try to bolt on afterward.

The minimal version of a retention system looks like this. Before you open a book, write down one question you want the book to answer. While you read, take brief notes in your own words — a sentence per major idea is enough. After you finish a chapter or a summary, spend two minutes trying to recall the key points without looking at your notes. And use a tool that handles the spaced repetition scheduling for you, because doing it manually is how good systems die.

Sumizeit handles the last piece natively. The flashcard system runs automatically after every summary, schedules your reviews without you having to track anything, and fits into the gaps in your day — two minutes on the commute, three minutes in a waiting room — rather than requiring a dedicated session. The summary itself takes 15 minutes. The review that makes it stick takes a few minutes more. The total investment is under 20 minutes for a level of engagement with a book's ideas that will be accessible to you six months from now.

That is a fundamentally different relationship with reading than most people have. And it starts with understanding that getting through books is not the goal. Getting ideas from books into your brain, where they can actually do something, is the goal. Everything else is just a pile of dog-eared pages.


The Reader You Want to Be Is a System Away

The version of you that confidently recalls what you read, references books fluently in conversation, and builds on ideas across multiple books over time is not separated from you by a memory you don't have. It is separated from you by a reading system you haven't built yet.

The system does not require hours of daily study or a perfectly organized notes library or a complicated productivity setup. It requires reading with a question, noting ideas in your own words, and revisiting what you've learned at spaced intervals. Sumizeit's flashcard feature automates the most technically demanding part of that — the scheduling — so that the system actually runs instead of quietly collapsing under its own friction.

The forgetting curve is real. But it is not a law of nature that has to apply to your reading life. It is a default that you can override, one review card at a time.

Start with a summary. Generate the flashcards. Come back to them when Sumizeit tells you to. Do it again with the next book. In three months you will have a relationship with the ideas you've read that you have never had before — and you will understand, finally, what reading for retention actually feels like.


For 15-minute non-fiction book summaries of best-selling books, check out sumizeit.com

Great books in a fraction of the time

Get the key insights from top nonfiction books in text, audio, and video format in less than 15 minutes.

Get 2 FREE sample summaries!