The Prompt Was Too Big

On June 2, I did something that now feels both brave and slightly ridiculous.

I opened the AI builder and asked it to make the entire Beads of Hope T1D website in one giant prompt.

Not a homepage. Not a test page. Not “let’s start small and see what happens,” which probably would have been the sensible thing.

I asked for a “fully functional, production-ready ecommerce and community website” for a philanthropic organization called Beads of Hope. I told it the purpose: raising money and awareness for Type 1 Diabetes through handmade, custom-designed bead products.

Then I basically handed it the whole dream in one breath.

Shop. Custom products. Community. Checkout. The works.

I am 50, with no coding background, so there was a part of me that expected the AI to come back with something that looked like a school project. Maybe a pretty front page. Maybe some pretend buttons. Something I could point at and say, “Well, that’s a start.”

About 20 minutes later, it had built the bones of a real site.

There was a shop page. There was Google login. There was Stripe checkout connected in the structure of the app. There were 22 database tables, which I only sort of understood as “places where the website stores information.” Products, orders, users, carts — all the invisible filing cabinets a real online store needs.

And then I saw the storefront.

That first moment was honestly thrilling. It looked real. Not finished, not perfect, but real enough that I had that strange little stomach drop of, “Oh. This might actually happen.”

It was also completely disorienting.

Because when something suddenly looks real, your brain starts asking real questions. Can people buy from this? Is it safe? What if I break it? What if someone clicks something and it does the wrong thing? What if I don’t understand enough to know what’s wrong?

I went from imagining a website to staring at one, and that shift was bigger than I expected.

Of course, the magic did not last long before the first wall appeared.

That same evening, none of the uploaded product images were visible on the published site. The shop page also had duplicate category buttons, and every time I clicked where a product image should have been, the screen snapped back to the top of the “all” category.

I remember telling the AI, “images are still not visible and each time I click on the box the image should be in, the screen immediately reverts back to the top.”

That sentence pretty much sums up my first long night with the project: something amazing existed, but I couldn’t make the pictures show up.

And the AI would fix one thing, then another thing would still be weird, and I would try again. Hours went by like that. I was excited, but also annoyed and tired and starting to wonder if I had asked for too much too fast.

Looking back, I probably did ask for too much in one prompt.

But I’m not sure I regret it.

That first big prompt gave me the whole shape of the idea. It turned Beads of Hope from something in my head into something I could click through. Even with the broken images and jumpy page and duplicate buttons, I could finally see the store I was trying to build.

It did not live up to my expectations in the sense of being instantly done. That was the fantasy version. The real version was messier.

But it exceeded my expectations in a different way. It showed me that I could start with plain English and end up with a working foundation for something that mattered to me.

The first big prompt did not finish the project.

It made the project real enough that I couldn’t pretend it was just an idea anymore.

Small takeaway: asking for everything at once may not be the cleanest way to build, but sometimes it gives you the courage to begin.