When I started designing this thing back in early 2015, I didn't sketch a design tool. I sketched a stack. Elements piled on top of each other, top to bottom, and that was the whole idea. People kept asking me why I didn't just make it more like a Canva landing page builder, a big open canvas, drag anything anywhere, do whatever you want. And every time, my answer was the same: because that is the part I wanted to cut.
So let me walk through the reasoning, because it wasn't obvious even to me at the time.
Why not just build a Canva-style landing page builder?
Here's the honest version. I'd spent years designing and coding site templates, then building whole sites to hand those templates out. It was good work. It also got routine, the kind of routine where you know you should try something new but you're not sure what. When I finally decided to branch out, the open-canvas approach was the tempting one. More power. More freedom. Let people move any box anywhere.
But a blank canvas is a lot of rope. For a non-technical person who just wants a one-page profile or a small landing site fast, all that freedom is the problem, not the feature. You open it up and you're staring at nothing, making a hundred tiny layout decisions before you've written a word.
So I made a call and committed to it: don't try to do too much. Limiting the focus is the single thing that made the whole product feasible for one person to build, and, as it turned out, the thing that made it easy to use.
What does "a simple stack of elements" actually mean?
Instead of a canvas, you get a stack. You add an element, a heading, an image, a button, a form, and it drops onto the pile. Order it, style it, done. There's a natural limit to how complicated that can get, and the limit is the point.
Think about what that removes. No wrestling with layers. No aligning boxes by eye. No "why is this element floating in the corner." The structure does the hard part for you, so the only decisions left are the ones you actually care about, the words, the colors, the details.
That's the trade a Canva landing page builder can't quite make, because it's built to do everything. Mine is built to do one thing. Simply built. Customizable where it counts. Responsive out of the box, and I mean a mobile-friendly builder, not just mobile-friendly sites, because you should be able to put one of these together from your phone.
How do you fit deep customization into a narrow tool?
This is the part I worried about most. Narrow is great until it becomes limiting, and then people bounce. All looked great on paper, but how would it play out in practice?
The answer was to be strict about the shape and generous about the details. The stack pattern stays fixed, that's non-negotiable, that's what keeps it simple. But inside that structure, you can go deep: fine-grained control over spacing, typography, colors, backgrounds, the stuff that separates a site that looks templated from one that looks like yours.
I also left room to be wrong. I stuck to the vision but kept a little flexibility, and I'm glad I did. Someone asked for multiple pages, which sounds like it breaks the whole one-page premise, and instead of saying no, it became the Control Element. The narrow focus held. The tool just got a bit more capable inside its own rules.
Why break it into five pieces?
Carrd was a big project by my past standards, big enough to be intimidating if I stared at the whole thing at once. So I broke it into five subprojects, Generator, Builder, Dashboard, Site, and the Paid Plan, and took them one at a time.
That wasn't just project management. Each piece got to be small enough to actually finish. The Builder is where the stack lives. I had a rough sense of how it should look from that 2015 conception. I'd even done a mockup before I wrote a line of the Generator, and after a few months of iterating, the thing landed remarkably close to that first sketch.
The Dashboard I built around big auto-generated screenshots of your own sites, so the hub is your work staring back at you, not a wall of filenames. Same instinct as the Builder: make the thing you made the centerpiece.
The unglamorous engineering call
Here's one I still get questions about. I built the Builder framework-less, vanilla JS and jQuery, hand-rolled. In 2015 that already felt a little stubborn, and today it sounds downright contrarian.
But going without a framework meant the tooling fit the Builder instead of the Builder bending around the tooling. I knew exactly how every piece worked, which made fixing bugs and adding features far less mysterious. And it forced me to learn a pile of things a framework would have quietly hidden from me. Would I recommend it for everything? No. Worth doing at least once? Absolutely.
What about the name?
Small detail, weirdly hard. The name would show up inside people's own subdomain URLs, so it had to be short, memorable, and low-key enough not to stick out in the middle of someone else's address. Then reality: domain availability across those trendy two-letter extensions is a brutal constraint. "Carrd" at carrd.co threaded the needle, short, sticks in your head, disappears politely into a URL.
Would I do the narrow thing again?
Yes. Without hedging. The lessons all pointed the same direction. Don't fixate on competing products, or you'll box yourself in copying features you never needed. Don't force people to sign up just to try the thing, let them poke at it first. Plan away from the screen, on paper and whiteboard, where it's easier to think. Always test on real devices, never just emulators. Lean on communities like Twitter and Product Hunt for honest feedback. And, genuinely, have fun with it.
But the biggest one is the boring one: pick a niche that's simple, has wide appeal, and is already proven popular, then stay inside it. A Canva landing page builder wins by being able to make anything. This wins by making one kind of thing so well that a total beginner can crank out a gorgeous site in under 5 minutes. Different bets. I'm happy with mine.
Narrow isn't the apology here. Narrow is the whole product.