All writing

Nobody deletes an app because of its framework

A friend asked whether to build native or use Flutter. Why AI may bring big companies back to native apps, and why a new product should still ship first.

A friend called me with a new app idea. He was excited, but stuck on the very first question: "Should I build it native, or use something like Flutter?"

He had read all the debates. Native apps are fast and feel right. Cross-platform apps are cheaper, but "not real apps". He wanted to get it right from day one.

I asked him one question back: "How many users do you have?"

Zero, of course.

Why one codebase won

Imagine you want your product on iOS, Android, Mac, Windows, and Linux. Going fully native means five codebases, often five teams, and five ways to build the same button. Every feature ships five times. Every bug gets fixed five times. Release dates drift apart, and the cost keeps climbing.

Frameworks like React Native, Flutter, and Electron came to solve exactly that. One codebase, many platforms, one small team.

The price is that the app is not truly native. Electron carries a whole web browser inside it. Flutter draws its own pixels. React Native maps your code to native parts through an extra layer. Most users never notice, but some do: a slow start, too much memory, scrolling that feels a little off, a menu that doesn't behave like the rest of the system.

AI changes the math

Here is what I believe: in the AI era, many big companies will go back to native apps.

The main reason to share one codebase was cost. Writing the same feature in Swift, Kotlin, and C# took people, time, and a lot of meetings. Now an engineer can build a feature on one platform and ask AI to port it to the others. AI can read a Swift screen and write the Kotlin version, keep the tests in sync, and explain the platform rules you haven't learned yet. You still need someone who knows what good looks like on each platform, but the work that used to need five teams can be done by a much smaller one.

When the cost of native drops, its benefits start to look very attractive again.

So are the frameworks dead?

No. And this is where I told my friend to relax.

Speed and polish matter when people are already using your product. Before that, the only thing that matters is whether anyone wants it at all. Building something truly helpful is far more difficult than any technical choice. I have never seen a product fail because it picked the wrong framework. I have seen plenty fail because nobody needed them.

So pick the tool that gets your product into real hands fastest. For most new products, that is still one codebase. Less is more: one codebase, one small team, one goal.

When your users start complaining that the app feels slow, celebrate. That is a great problem to have. It means you have users. Now you've earned the right to go native, and AI will make that move cheaper than ever.

Nobody deletes an app because it was built with Flutter. They delete it because it didn't help them.

Ship first. Go native when your users give you a reason.

Keep reading

More from the notebook.

AI2 min read

Phishing in minutes, not weeks

A fake bank text that almost got me, a deepfake video call that cost a company millions, and why the answer to AI-powered scams is to get sharper, not to stop AI.

Building3 min read

Praying to the servers

An index that froze the orders table, a shrine for praying while we waited, and why the best server prayer is the one nobody has to say.