Stop Waiting Until It’s Ready

There’s no such thing as “ready.” There’s only “good enough to start learning.”

In the early days of Netflix, I was a perfectionist.

Not the admirable kind — the paralyzing kind. The kind where you spend a month commissioning custom photography for a test that ultimately fails in the first hour. The kind where you labor over every word of copy and stress-test every link and hire a copy editor before a single real customer has even seen the page.

You know what all that careful preparation got us? Slower.

Every test took weeks. Every test failed. So we’d start over, and do it again. Weeks. Fail. Weeks. Fail.

Finally, out of desperation more than wisdom, we started cutting corners. We did a test in two weeks. Then one week. Then every other day. Then every day. And then — because we had no choice — three or four different tests a day.

Things got sloppy. Photos still had watermarks. There were typos, dead links. We crashed the site. It was a mess.

And here’s what we discovered: it didn’t matter.

Not only did our sloppiness not hurt us — our speed helped us enormously. The insights we gathered in a single sloppy week of testing were worth more than a month of perfect preparation. We stopped trying to get it right in advance and started letting reality tell us what “right” actually looked like.

That’s when we stopped being a company that made careful decisions and became something far more valuable: an idea machine.

The perfectionism trap

Here’s why perfectionism is so insidious. It doesn’t feel like fear. It feels like quality control. It whispers that you’re being thorough. Responsible. That you’re just waiting until things are truly ready.

But there’s a word for building something in your head and refusing to show it to the world until it’s perfect.

That word is stalling.

The sculptor analogy I always come back to: when you’re doing iterative testing, you have to get used to having your perfect sculptures squished. You spend two weeks crafting something beautiful — the right messaging, the right design, the right flow — and then you find out it doesn’t work. Squish. Start over.

If you can’t handle that, you’ll find yourself building bigger and bigger sculptures, taking more and more time, spending more and more money — all in service of delaying the moment when reality gets to weigh in.

Perfectionism assumes you know what “perfect” looks like in advance. You don’t. I didn’t. Nobody does.

No business plan ever survives a collision with a real customer.

What we actually shipped

When we launched Netflix in April 1998, the website was clunky. The fulfillment system was manual and error-prone. Genuinely rough.

But it worked well enough that customers could find a movie, order it, and receive it in the mail. That was the bar. Everything else was secondary.

And because we launched when we did — instead of six months later, polishing things that didn’t need polishing — we found out stuff we never could have learned in a conference room. That late fees were a bigger emotional trigger than we’d realized. That people were more willing to wait for a DVD than conventional wisdom suggested. That our packaging needed serious work.

None of that was available to us until something real was in front of real customers. Focus groups don’t tell you what people will actually do. Behavior does.

You’ll learn more in one day of doing than in a month of thinking about it.

The actual cost of waiting

What perfectionists never calculate: every day you don’t launch is a day you don’t learn. And in a startup, that’s not a small thing.

The subscription model at Netflix — probably the single most important decision we ever made — wasn’t the result of careful analysis. Reed and I were in the warehouse one day, looking at 100,000 DVDs just sitting on the shelf. We were almost out of money. We were joking that maybe we could save on rent if we could store the DVDs at customers’ houses.

And then we thought… what if we just mailed them and let people keep them as long as they want? No due dates. No late fees. Just a flat monthly subscription that let them exchange discs as often as they wanted.

It wasn’t a carefully researched strategic pivot. It was a half-baked idea we decided to try. And astoundingly enough — when we tried it, it worked.

Perfect information doesn’t exist. You’re always making a bet with incomplete data. The question is whether you make that bet early, when the stakes are small, or late, when you’ve burned months and money protecting an idea that’s never been tested.

What “good enough” actually means

It doesn’t mean everything works. It doesn’t mean you’re not embarrassed.

“Good enough” means one thing: it’s the version that lets you start learning. Not the version that covers every edge case or impresses your board or makes you feel confident at a dinner party. Just enough to find out if there’s a there there.

The test I always apply: what’s the quickest, cheapest, easiest way to try this? Not the best way. Not the right way. The fastest way to find out if the idea has legs.

If it does — great. Now you can invest more. If it doesn’t — you’ve lost a week, not a year.

Build the thing. Ship it. Let reality sort it out.