We built a full online store for a collectibles card shop, Tarjetas Coleccionables México. Real catalog, real product pages, real photography, a clean way to browse thousands of cards. And it takes no payment online at all. No cart wired to Stripe. No card fields. Nowhere on the site can you actually pay.
That reads like a corner we cut. It wasn't. It was the right call, and we'd make it again tomorrow.
The default is not neutral
The owner's customers already buy a certain way. A message. A back-and-forth about condition and price. A handshake over WhatsApp. That motion already works. It closes sales every day, and it did so long before any website existed.
Now picture bolting a checkout onto that. A payment processor. Its fees on every sale. Its fraud surface. A returns policy someone now has to write and stand behind. All of that machinery would add friction to a motion that was already frictionless. Worse, it would move the sale out of the channel where it already closes and into a channel nobody asked for.
Here's the part most studios miss. The standard ecommerce default (add to cart, enter card, done) is not a law of nature. It's a decision. Software's job is to fit how a business already works, not to impose a template because the template is what everybody ships. For a small operator, the default is often the wrong choice, and shipping it anyway is laziness dressed up as completeness.
A checkout you don't need isn't a feature. It's a tollbooth you built on your own road.
What we built instead
So we followed the real motion. On the site you browse the catalog, add the cards you want to a list, and see prices as you go. When you're ready, one tap drops you into the owner's WhatsApp, with your whole list and the prices already filled in. The conversation starts where it always started, except now it starts with a clean order instead of a cold "do you have this?"
The close happens where it already happened. The website does the tedious part (the browsing, the finding, the tallying) and then hands the human back to the human. Zero checkout, and the owner didn't lose a single thing he had before. He gained a catalog that sells for him around the clock.
The wider lesson
The best feature is often the one you refuse to build. A checkout nobody asked for isn't neutral overhead. It's pure liability with no upside attached:
- Payment compliance you now own, forever.
- Chargebacks and disputes that used to be someone else's problem.
- Abandoned carts: a brand-new way to lose a sale you would have closed by message.
- A returns policy you have to write, publish, and honor.
- A processor's cut on every transaction, for the privilege of all of the above.
Match the real motion of the business. Add the machinery only when it removes friction, never to look finished. A store that looks "complete" and quietly bleeds sales into the gap between the demo and real life is worse than a store that's honest about how its customers actually buy.
None of this is dogma
To be clear, we're not against checkout. When a business genuinely wants self-serve payment (real volume, no human in the loop, customers who expect to pay on the spot), we build that, and we build it well. One of our own demos, a spirits storefront, has a full cart and a real checkout, because that business calls for it.
That's the whole point. Checkout is a choice you make per business, weighed against how its customers actually behave. It is not a default you spray across every store because the tooling makes it easy. Fit the software to the motion. Sometimes the motion has a card reader in it. Sometimes it has a phone number.
The cart was never the point. The sale is. And sometimes the shortest line between a business and a sale skips the cart entirely.
Questions are welcome: hola@xala.studio