At one point I had a streak going. 47 books in a year. I tracked them in a Notion database, tagged by category, rated by usefulness. I felt productive. I felt like I was growing.
Then someone asked me to explain something I'd read three months earlier and I basically just stared at them.
I could remember the book's cover. I remembered thinking "this is good" while reading it. But the actual ideas? Gone.
There's a specific kind of developer who reads a lot. Maybe you know them, maybe you are them. They've read Clean Code, The Pragmatic Programmer, Refactoring. They can recommend books confidently. They talk about principles fluently.
But their code looks the same as it did two years ago.
I was that person for a while. Reading felt like learning because it has the same surface texture as learning. You sit down, you focus, you process information. Your brain feels tired afterward. That tiredness gets misread as progress.
It isn't.
It took me a while to figure out what was actually broken.
Think about it this way. You want to make pasta but you've never done it before. You're standing in the kitchen, stuck. Then you spot a recipe book on the shelf.
Here's how most people open that book: they flip to the pasta section, start reading, absorb as much as they can, maybe flag a few pages, close it, and then try to remember what they just read while standing back at the stove.
Here's what works better: before you open the book, write down the question you actually need answered. Not "learn about pasta." Something specific. What ingredients go into a basic pasta dish? One question. Concrete. Driven by the thing you're already trying to do.
Then you open the book. You read until you can answer that question. When you can, you close the book and you go make the pasta.
That's it. That's the whole method.
The sequence matters more than it looks.
The question comes first — before you open anything. Not after skimming the introduction. Not while you're already three chapters in. Before. Because the question is what turns reading into a search instead of a scroll.
Without a question, you're a tourist. You notice interesting things, take a few photos, move on. Nothing changes when you get home.
With a question, you're looking for something specific. The book stops being content to consume and becomes a place to find an answer you already need. The difference in retention is not subtle.
I started applying this to technical books.
Before opening a chapter on dependency injection, I'd write: Why does my ViewController feel impossible to test? That's the question I actually had. That's the thing I was already stuck on.
Then I'd read. Not the whole chapter necessarily — just until I had enough to attempt an answer. Then I'd close the book and go refactor something.
The refactor was always ugly. I never fully understood what I was doing the first time. But two weeks later I remembered everything, because the memory wasn't just "I read about this." It was "I had a question, I found a partial answer, I tried it, it was messier than expected." That kind of memory has texture. It survives.
The chapter alone would have evaporated. The question plus the chapter plus the attempt stayed.
What tripped me up for years was mistaking fluency for understanding.
After finishing a book I could talk about it. I could summarize the argument. If someone brought it up I had something to say. That felt like learning. But being able to talk about an idea and being able to use an idea are completely different cognitive operations.
I could describe what a design pattern does. I could not reliably recognize when to reach for one.
The question-first method closes that gap because it forces the connection to exist before you even start reading. You're not trying to find a use for knowledge after the fact. You already have the use. You're just looking for the knowledge to fill it.
Now I read around 100 books a year. The reading volume went up and the actual skill level went up with it.
Because every book now has a question attached to it before I open it, and something I tried after I closed it. Reading and doing stopped being separate activities. They became the same loop.
Reading is how you find out what exists, what's possible, what other people have already figured out. That part is genuinely valuable.
But a recipe book doesn't teach you to cook. Standing in the kitchen, failing to make pasta, then picking up the recipe book with a specific question — that teaches you to cook.
Don't open the book first. Write the question first.
Then go make the pasta.
Originally published at devkoan.substack.com
