We ran a website service by hand for a year. Here's what it taught us.

Pages started with a bill I couldn't pay.
An electrician, McEvoy Electric, had done work for me. The invoice came to six hundred dollars I didn't have that month. So I did what a wolf does. I read the terrain. I pulled up his website, and it was tired, the kind of site quietly costing him customers he'd never know he lost. In under an hour, I built him a new one. Then I asked how much he'd take off my bill for it.
He took all of it. Six hundred dollars, gone, traded for a website that took less time than a service call. That was the moment I understood what AI had actually changed. It's the moment Pages comes from.
A word on the wolf, because the path here matters. I'm Josh Wolf, two decades a working journalist, an M.J. from Berkeley, a Logan Prize, trained in one skill above all others: cut through the noise, find the fact that matters, say it in language anyone can read. In Pickaxe's earliest days I was the company's evangelist, out front telling people what no-code AI could actually do. Then I stepped back to go prove it myself. I opened a small AI practice in Nevada City called IntelliBotique, on one promise. Send the wolf, get it handled. One practitioner. No middlemen. No farmed-out drafts. A wolf reads the terrain, moves, and does the work.
The electrician wasn't a one-off. The same problem kept walking in. Good businesses, bad websites. Not bad people, not bad work. A site from three rebrands ago on a broken theme, quietly undercutting everything they'd built. So I started fixing them on purpose, on my own dime, under my own name. It began as one stubborn promise, the one-hour website, and grew into a real service: SiteCrafters. Your site, refreshed, in about an hour. My shop, my guarantee, nobody's permission but the client's.
The hour wasn't a gimmick. In 1983 a man named Dean Butler asked why new glasses had to take weeks, opened a store that did it in an hour, and had 350 locations in five years. LensCrafters. We had the same thought about websites. Why does a refresh take six weeks and ten thousand dollars? It doesn't. We don't start from scratch. We diagnose. We prescribe. We craft. We see what's there and make it worth seeing.
So I did it, by hand, live, while clients watched. The electrician. A studio. A fire-safety nonprofit. Flat rate from $399, and a guarantee that still stands: if it's not better, you don't pay.
A year of that taught me four things. They're the reason Pages exists.
The bottleneck was never the building. Once I knew what a site needed to say, building it was fast. The slow, valuable part was the seeing, reading a business closely enough to find the one true thing it had buried. That was the craft. The HTML was the easy part.
Every site had the same disease. The homepage said nothing only that business could say. McEvoy Electric, licensed, twenty-plus years in Nevada County, had a site that could have belonged to any electrician anywhere. The fix wasn't a prettier template. It was moving the license and the decades above the fold, where a stranger sees them first. Same business. I just stopped hiding the proof.
The owner always knew the answer. Every time, on the first call, I asked "why should someone pick you?" Every time the answer was sharp, specific, and missing from the website. The raw material was never gone. It just never made the page.
And the method repeated. Audit honestly. Study the best. Keep, rewrite, cut. Find the one true thing. Show real directions. Write it in the owner's voice. Build it clean, on files they own. By the tenth client it wasn't instinct anymore. It was a process. The same eight steps, every time.
That's the moment a service becomes a product. When you've done a thing by hand enough times to see the pattern, and the pattern holds, you can hand it to software. Not to replace the judgment. To do the ninety percent that was always mechanical, fast and cheap, so the judgment is all that's left to weigh.
Which is the moment I came back to Pickaxe. I'd helped tell its story at the start. Now I had a method worth automating, and Pickaxe was the no-code AI workshop built to automate exactly this kind of thing, the platform I'd spent years championing before I ever ran a refresh by hand. So the project came home to Pickaxe, and the one-hour website I'd been running solo became Pickaxe Pages.
That's what Pages is. The SiteCrafters method, productized with Pickaxe. Paste your URL, the engine runs the same diagnosis and rebuild I used to run live, and you approve the result. No booked session. No waiting an hour. No watching me work. The craft is still in there. We just taught it to run on its own, on rails I trust, because I helped make the case for them.
I'm telling you the human origin because it matters who's behind a tool, and because the honesty is the point. Pages isn't an AI gimmick guessing at what a website should be. It's a year of real businesses, real engagements, a guarantee I backed with my own name, and a method good enough to bring to Pickaxe and turn into something you can run yourself in minutes.
Next, the eight steps in plain language, so you can see the method whether you ever touch the tool or not.
software didn't invent this method. A year of doing it by hand did. And the hardest part, the seeing, is the part we worked hardest to teach it.