“If you’re not embarrassed by the first version of your product, you’ve launched too late”
— Reid Hoffman, LinkedIn founder

Most founders are very focused on their idea. But when you look at ventures that have actually been successful, they often look completely different to what they were first intended to be. In fact, there are lots of familiar examples where the idea that eventually worked is so different as to be unrecognisable.

In 2000, when Trade Me was free, the tagline we used on the first billboards was: “Only turkeys pay for classifieds”.

(The idea that Trade Me didn’t waste/spend any money on traditional marketing like billboards is another myth, unfortunately.)

In 2000 at Trade Me, we thought free online classifieds would create a business opportunity to sell online advertising. It turns out we were the turkeys for thinking this business model was the best option.

This is what the Trade Me home page looked like when I first started working on the site in early 2000.

We thought it was great—including the Wingdings inspired logo. Now it’s a little embarrassing. But that’s the point …

The only way you learn which parts you need to be embarrassed about later is by launching and listening carefully to the feedback you get.

You need to listen to what people say, but most of all pay close attention to what they actually do. An empirical approach like this makes you humble. You quickly learn what little things you got right and what big things you got wrong.

So, don’t try to partially complete the whole product: complete a part of the product (preferably the part that will demonstrate value to paying customers). Then extrapolate from there based on how people actually use it. It’s nearly always better to implement a smaller number of features really well than trying to do everything averagely.

If you get this right over time then you too can look back and retrospectively invent a story about the mythical magical moment when the idea popped into your head, perfect and fully formed.