Why we write the homepage before we design it

[Invented] Most homepages start with a layout and end with copy squeezed into it. A headline gets trimmed to fit a box. A paragraph gets cut because the grid ran out of room. By the time anyone reads the words properly, the design has already decided what they're allowed to say.

[Invented] We try to do it the other way around. Before we open a design file, we write the page. Plain text, in a document, top to bottom. What does a visitor need to hear first? What do they need to believe before they'll scroll? What should they do at the end?

[Invented] It's a humbling exercise. Founders arrive with a lot to say and no order to say it in. Writing the page forces the order. You find out quickly which claims you can actually back up and which ones were just habit.

[Invented] The first draft is usually bad. That's useful. We read it out loud, in the room, with the client. Anything that makes someone wince gets cut. Anything that makes someone say "yes, exactly that" gets moved higher up the page.

[Invented] There's a test we use. Cover the logo and the product name, and read the page to someone who's never heard of the company. If they can't tell you what it does and who it's for by the end of the first screen, the words aren't done. No amount of design will fix that.

[Invented] Once the words hold up, design gets easier. Not because we've got less to do, but because every decision has a reason. The headline is big because it's the point. The button is loud because the page has been building towards it. Nothing is decoration, because the copy has already told us what matters.

[Invented] It also changes how we talk to clients about feedback. Instead of "I don't like the blue", the conversation is about whether the page says the right thing in the right order. That's a better argument to have, and it's one we can usually settle in a single call.

[Invented] It doesn't mean the copy is locked. Design pushes back sometimes. A section turns out too long. A line works better as a label than a paragraph. We edit as we go. But we edit from a page that already makes sense, rather than patching one that never did.

[Invented] Founders sometimes worry it slows things down. In our experience it does the opposite. The weeks we spend circling a layout we can't commit to are the weeks we haven't decided what the page is for. Writing first moves that decision to the start, where it's cheap.

[Invented] We learned this the slow way, by designing beautiful pages around sentences nobody believed. Nobody complained about the layouts. They just didn't convert, and nobody could say why. The copy was the why.

[Invented] A note on tone. Write like you talk. If you wouldn't say a sentence across a table, it doesn't belong on the page. Most of the stiff, corporate copy we see was written for an imaginary reader. Write for a real one.

[Invented] If you're building a site yourself, try it. Open a blank document, write the page as if you were explaining your company to a friend, and only then open Figma. You'll probably throw half of it away. The half you keep is the half that matters.