Before writing a single line of code, I made a plan to ask real people first

I had an idea, someone keen to work on it, and the skills to build it.

So obviously I decided not to build it.

Not yet, anyway. Not until 10 to 15 people on each side told me it was worth it.

The idea was an ordering app for local shops. Customers browse, shops receive orders, everyone saves a trip. A collaborator had sent me a first summary a few days earlier. It was good. I sent back a simpler version, a rough mockup on screen and, more importantly, a list of things to do before anyone opened a code editor.

Some basic explanatory material first. A short presentation, a flyer. How it works, and no prices, so people react to the idea and not to a number.

Then two surveys. One for the shops. One for the people who would buy through the platform.

Ask 10 to 15 of each. Read the answers honestly. If they said no, we'd stop, or pivot. We even had a spare angle in the drawer, centralising menus, in case the first one didn't hold.

Only if the answers were good would I start programming. I also proposed how to split the work between us: the material, the surveys and the asking itself.

Where I like to hide

This plan wasn't wisdom. It was fear of a mistake I already knew.

I build websites and tools for a living. Building is the part I enjoy. It's also the part where I can hide.

When you're coding, you feel productive. Every finished screen looks like progress, the same way a tidy desk looks like work. But a finished screen tells you nothing about whether anyone will use it. Or pay for it.

Code is the most expensive way to find out nobody wants your idea.

That's why the surveys only had two jobs.

Would both sides actually use it? A shop and a customer both have to change a habit. If only one of them wants it, there's no product. Just a very nice screen.

And would anyone pay? Liking an idea costs nothing. Paying for it is the only vote that counts.

Everything else, the colours, the features, the name, can wait. The name especially. I suspect more side projects have died arguing about a name than about a price.

Ten to fifteen people isn't statistics. But it's enough to hear the same objection three times. And the moment to decide what "no" looks like is before you read the answers, not after. Otherwise you'll read whatever you want into them.

Patient, not infinite

Skipping all this feels faster. You get to the fun part sooner.

But nobody funds us. We started Foxnice with 300 €, and every product we build is paid for by client work. Every hour on our own product is an hour no client paid for.

That money is patient. It isn't infinite.

A week of surveys costs almost nothing. Three months building the wrong thing costs a season of client hours. And when you've asked people first, you start building with their words in your head, not just yours.

I could have started coding that same day.

Instead I'm going to spend the next few weeks asking. If the answer is no, I'll have lost a few conversations, not a product.

P.S. For the record, the editor is still closed. It's killing me.