Most of what we read, we forget.
This is the uncomfortable open secret of readers, and an entire industry has been built on solving it. There are note-taking apps with cult followings. There are systems — Zettelkasten, Building a Second Brain, the commonplace book revival — promising that if you only capture, link, and review what you read, your reading will finally compound into something. There are YouTubers who have made careers explaining how to take better notes on books. There are spaced-repetition tools that will surface a sentence you highlighted in 2019 on a Tuesday morning in 2026, so you can be reminded that you once thought it was important.
All of this is built on a premise the industry rarely states out loud: that the value of reading is what you can retain.
I think the premise is wrong, and I think it has quietly distorted what an entire generation of readers thinks they’re doing when they sit down with a book.
The retention trap
Try a small experiment. Pick a book you read more than two years ago and would still recommend — a book you would say, without hesitation, changed how you think. Now, without checking, write down its core argument in three sentences. Write down two specific examples it used. Write down one line that stuck with you.
Most readers, doing this honestly, find they cannot. The argument is gone. The examples are gone. There may be a single line, or a single image, or there may be nothing at all. And yet they still think the book changed them.
Either they’re wrong about the book, or they’re wrong about what changed means.
I think they’re right about the book and wrong about retention. The book did change them. The retention industry has just sold them a definition of “change” that’s narrower than the thing they actually experienced.
The narrow definition is something like: a book has changed you if you can later quote it, summarise it, or apply its specific frameworks. The broader definition — the one that matches how reading actually seems to work — is closer to: a book has changed you if your thinking, after reading it, is different from your thinking before, even when you cannot reconstruct the path between the two.
Retention is one route to change. It is not the only route, and for most readers, it is not the main one.
What the books actually do
I read Thinking, Fast and Slow eight or nine years ago. If you asked me to explain Kahneman’s dual-process theory, I could give you a rough version. System 1 is fast and intuitive. System 2 is slow and deliberate. Beyond that, the specifics begin to blur. The biases — anchoring, availability, the endowment effect, prospect theory — I can name some of them. I cannot give you the experiments. I cannot remember which chapter argued what.
By the retention industry’s standard, this is a failure. I read a book, I should have notes on it, the notes should be searchable, and they aren’t.
By any honest standard, that book is one of the half-dozen most important books I have ever read. I think about my own thinking differently because of it. I notice when I am pattern-matching. I am suspicious of my own confidence. When someone presents me with a vivid example, I have a small alarm that goes off — availability bias — even when I can no longer remember which page that term appeared on. The book is not in my notes. It is in my reflexes.
This is, I suspect, how most serious reading actually works. The book gets metabolised. The specific arguments dissolve, the way the specific words of a song dissolve into the song itself. What remains is not the content but the orientation — a slight tilt in how you see things, a new frequency you can now hear.
The retention systems are designed for a kind of reading where this isn’t enough. They are designed for academics, lawyers, knowledge workers who need to cite what they’ve read. For that work, retention systems make sense. If your job is to assemble arguments out of other people’s arguments, you need a way to find them again.
But this isn’t most reading. Most reading is not citation work. Most reading is the slow shaping of a person by the books they let inside, and that shaping does not require recall. It requires only that the book was good, that you read it carefully enough at the time, and that you didn’t fight what it was trying to do.
What the systems are actually solving for
If retention isn’t the goal, why are the retention systems so popular?
Two reasons, both worth being honest about.
The first is that retention is measurable and effect is not. You can count notes. You can review highlights. You can see your Readwise streak. You cannot see how a book changed you — that’s a thing you can only notice years later, in moments where you behave differently than you would have without the book. The systems are popular partly because they give you something to point to. Look, I read this book. Here are my notes. The notes feel like proof that the reading happened. The change is invisible.
The second reason is that retention systems offer the comfort of activity. Highlighting, tagging, linking, reviewing — these things feel like work. They feel productive. They make reading feel less like a soft pleasure and more like a discipline you can document. For people who feel guilty about reading for its own sake — and there are many such people, especially among the kind who buy productivity tools — the retention system gives reading a respectable cover. I’m not just reading. I’m building a knowledge base.
I am not against discipline. I am against the substitution of discipline for the thing the discipline was supposed to serve. The retention systems began as a way to read better. They have become, for many of their users, a way to read worse — to read with one eye on the book and one eye on the note that needs to come out of it. This is a worse way to read.
How to read for effect
If retention is not the goal, what is the practice?
It is, mostly, getting out of the book’s way.
Read more carefully than the retention systems would have you read, in the sense of paying full attention, but read less defensively than they would have you read. Don’t underline as you go. Don’t pause to capture quotes. Don’t read with the half-formed thought that you’ll need to summarise this for your notes app later. The summarising posture changes how you read. It shifts you from receiver to processor, from someone the book is happening to into someone who is processing the book on behalf of a future self who will need the processed version. The book does its real work on receivers.
When you finish the book, sit with it for a few minutes. Don’t open the laptop. Don’t write anything. Notice what’s still there. Almost certainly, what’s still there is not the argument. It’s an image, or a feeling, or a sentence, or a question the book left you with. Whatever’s still there is what was alive in the book for you — and for your purposes, that’s the part that matters.
If you must take notes, take them sparingly, and take them later. A week after finishing, write down what you remember without checking. What you remember is what made it through. That’s the book’s actual gift to you. Everything else was scaffolding.
A small minority of books warrant something more — books you’ll teach from, books you’ll cite, books whose specific arguments you need to be able to reconstruct. For those, take real notes. But the test of whether a book is one of those is not whether it interested you. It’s whether you have a concrete future use for its specific content. Most books fail this test, and that’s fine. They were not bought to be cited. They were bought to be read.
What you actually have
The frustrating thing about reading for effect rather than retention is that you can’t show anyone what you have.
You cannot point to your bookshelf and say here is what these books did to me. You can only live in a way that’s been shaped by them, and let other people notice if they’re paying attention. The retention reader can show you their notes. The effect reader has only themselves.
This is, I think, the real reason the retention systems have won. Not because they produce better readers — they produce readers who are better at appearing to have read — but because they offer a tangible deliverable in a culture that has come to mistrust intangible work. We don’t know what to do with people who say a book changed them but can’t explain how. We are reassured by the ones with neat notes.
Don’t be reassured. The neat notes are the bookshelf, photographed. They are not the reader.
The reader is the person on the other side of the book — different now, in some way they can no longer fully account for, walking through the rest of their life with the book quietly working in them. That’s what reading is for. The forgetting is not a failure of the system. It’s the system working.
— Akhil M Sharma
