Ship It Ugly, Ship It Fast, Ship It Now

I used to be a perfectionist about product launches. Everything had to be polished. Every pixel considered. Every edge case handled. Every transition smooth.

That instinct served me well for years. Then AI showed up and made it a liability.

The speed gap

Here's what happened: the time it takes to go from idea to functional product dropped from months to days. In some cases, hours. And while I was still refining my color palette, someone else shipped a rough version of the same idea and started learning from real users.

They learned what worked. I learned what shade of blue looked best on hover states.

Guess who built the better product by month three?

Why ugly wins

An ugly product that exists beats a beautiful product that doesn't. This has always been true, but it's more true now than ever because the cost of iteration has collapsed.

When it takes three months to build something, you need to get it right the first time. You can't afford to ship garbage and redo it. So you over-plan, over-design, and over-engineer.

When it takes three days to build something, you can afford to get it wrong. You ship the ugly version. You see how people use it. You rebuild the parts that matter and throw away the parts that don't. The total cost of building, learning, and rebuilding is still less than the cost of trying to get it right the first time.

The perfection trap

I see this trap constantly with founders I work with through Nordic Intent. They have a great idea. They can see the finished product in their head. And they spend months trying to build that finished product before showing it to anyone.

By the time they launch, the market has moved. Someone else shipped a janky version that works well enough. Users don't care that the typography is inconsistent — they care that it solves their problem.

Perfection isn't a quality standard. It's a form of procrastination dressed up as professionalism.

What I mean by "ship it"

I'm not saying quality doesn't matter. It does — eventually. But quality applied too early is wasted effort.

Here's the order that works:

First, ship something that works. Ugly is fine. Incomplete is fine. It just needs to solve the core problem for someone.

Second, learn what matters. Watch how people use it. Listen to what they complain about. Notice what they ignore.

Third, make the important parts great. Now you know where to invest. Polish the things users actually care about. Cut the things they don't.

This sequence — ship, learn, polish — is dramatically faster than the alternative: plan, design, build, launch, hope.

Speed as a moat

In a world where AI gives everyone access to the same building tools, the advantage goes to whoever learns fastest. And you can't learn without shipping.

The founder who ships a rough prototype this week and gets feedback from ten users will build a better product than the founder who spends three months designing the perfect version in isolation.

Speed isn't about cutting corners. It's about getting to the truth faster. And the truth is always in the market, never in the Figma file.

The hard part

The hard part isn't building fast. AI has made building fast almost trivially easy.

The hard part is letting go of the version in your head. The polished, beautiful, perfect thing you imagined. You have to accept that the first version will be embarrassing. That people will see rough edges. That it won't feel like you.

Do it anyway. Ship it ugly. Ship it fast. Ship it now.

The beautiful version comes later — after you know what beautiful actually needs to look like.