All posts

Why we built Reytd

· Edward Karlsson · updated

For a few years, I worked at a web agency. We built online stores for our clients and looked after them once they were live. Most days were the usual agency mix: a new checkout here, a slow category page there, a client who needed something by Friday.

But every now and then, a ticket came in that I learned to recognise from its title alone: "Add product reviews."

That meant a review platform. Over the years I set up quite a few of them, from the biggest names in the business to smaller ones, and got to know them far better than I ever meant to.

The same story, every time

It always started well. The sales demo looked great, the client signed, and the work landed on my desk.

The first thing I looked for was somewhere to try it out: a developer environment, a sandbox, anything that wasn't the client's real store. Usually there wasn't one. So I built against the live account, tested on the real store, and hoped the review request I'd just sent off wasn't on its way to a real customer.

Then the script went onto the product pages, and the pages got heavier. We could spend weeks winning back speed on a store, and one review widget would undo a good part of that work.

And then the invoice came. That was the part these platforms were truly good at. Our clients paid a lot for them every month, and when I looked at what they got for it (stars, a list of reviews, an email after each order), I couldn't stop thinking that a small, quick system could do the same job.

"We could build that"

Every developer has said it about something. Usually it's a joke. This time, it stuck. I kept coming back to it, in the middle of yet another integration, and on the way home after it.

What would a review platform look like if it were built the way I'd wanted it, all those times?

  • It would be so light that nobody had to think about it on the page.
  • It would have a sandbox, so you could build and break things without a real customer ever noticing.
  • It would be documented properly, in the open, for whoever ends up doing the integration.
  • And it would cost what it's worth to a store, not as much as a store will pay.

So I started building it. That became Reytd.

Where it is now

Reytd opened to every store on 24 October 2026. It asks customers for a review once their whole order has been delivered, and shows the reviews on the store. The widget is about 9 KB compressed, and in our speed tests the stars show in less than 50 ms. Every store gets a sandbox with simulated emails, the docs are open to anyone, and the Free plan doesn't ask for a card.

It's still early. The Shopify and WooCommerce apps are on their way, and I have a long list of what comes after them. I'll write about it here as it happens, along with what I learn about reviews along the way.

If you've ever been the developer who got the "add product reviews" ticket, I built this for you. And if you run a store and you're tired of paying a lot for a little, have a look.