Everyone can ship things today. But most of what gets shipped cannot survive.

Survival is the real test. It's not about profitability. As time passes, a feature may no longer be used, a UI may look outdated, an essay may become nonsense.

In this essay I'll share what it takes to build things that last. I'm not talking about PMF or DAU. These have nothing to do with survival in this context, which is how the world judges whether a product sells.

Simplicity as a survival strategy

Universal entropy is naturally increasing.

An API can break. A dependency can get deprecated or become vulnerable. A link can go dead. To avoid the hassle of maintaining these ticking time bombs, the best approach is to make things simpler.

Consider OTP sign-in vs. password and OAuth sign-in. If your service needs an account basically for ownership verification, OTP wins. You don't need a Forget Password button. You don't need a Reset Password page. You just need to send that email.

Once part of your thing breaks, the rest will rot much sooner.

Simplicity doesn't mean less buttons. It's less cliche in the interface. Complex, feature-rich products like Photoshop and Linux have lasted decades. But they survive because of relentless maintenance and deep, irreplaceable value. If you don't have those resources, simplicity is your best bet.

Ask yourself before adding something:

• Will extra work be needed to keep this working if I switch to a new deploy platform or repo?
• Does this require yet another piece (like a new documentation page) to function?
• How likely is this to break or become deprecated within 3 years?

Simplicity does not mean faster progress. It means less surface area for decay.

Make things better, not more things

To make things survive, you constantly improve them. Digital things, like physical products, age. To slow down the aging, you remove unnecessary parts and keep maintaining the rest.

People quickly forget a hyped project. They may ignore what's already there, but they remember what's getting better. The accumulation of small improvements forges quality.

Make each word count. Optimize each image and video. Curate translation quality for each language. Polish every detail on each page.

The path to creating new stuff is unlimited — as is the path to making existing things better. This is your choice.

Some things are doomed to be buried. Don't build a patch for someone else's product unless you can always make it better than the official version. For example:

• An extension that exports pages from Notion — Notion will very likely support that natively.
• Feature add-ons for popular tools that are one update away from being obsolete.

Before you start, think: will this still be needed after 3 months? After 3 years?

Ship slowly

This sounds counter-intuitive: AI can speed up the shipping, why slow it down?

I like how Duolingo thinks about this:

You'll hear stories about a company going viral in a week. But the products that get big & stay big — they were patient. They weren't a short-term hack.

Luis

Their product didn't explode overnight. It compounded through years of patient, iterative improvement. That's the kind of durability worth aiming for.

Less hustle means less hype and more craft.

When you want to hurry something, that means you no longer care about it and want to get on to other things.

Robert M. Pirsig, Zen and the Art of Motorcycle Maintenance

This "blank time" is where inspiration grows. Value the time you feel "not ready" to work on the project. Do something else. Read a book, re-visit a movie, take a walk. The miracle is in there.

Notes

• Keep things simple, from internals to the surface.
• Make better things, not more things.
• Ship slowly.
• The Duolingo Handbook

No content is written with AI in this essay.