Rebuilding a Small Shop From Scratch

This is a case study about a small manufacturer. The name doesn’t matter for our purposes, so we’re not using it. What matters is that the project is the kind we do a lot of, for the kind of client we work best with, and the lessons it taught us apply to pretty much every project of this size.

The shop makes a physical product by hand. Letterpress printing, in this case, but it could just as easily have been ceramics, or custom woodworking, or small-batch textiles, or any one of a dozen other small-batch craft operations we’ve worked with. The pattern is similar across all of them. A talented founder. A small team. A product that’s genuinely good. A website that doesn’t come close to communicating any of it.

Here’s how we rebuilt it, and what we learned in the process.

What was broken

The founder came to us through a referral. We got on an intro call. Her first sentence was “I know I need a new website, I just don’t know what to do about it.”

That’s a sentence we hear often. What it usually means is that the current site has drifted a long way from what the business has become, and the founder knows it but can’t articulate the specifics.

In this case, the site was about four years old. It had been built on a theme by a previous developer who had since moved on. It had an about page, a shop with twelve products, a contact form, and a blog that hadn’t been updated in two years. The photography was okay but inconsistent. The copy was workmanlike. The brand was there in flashes but didn’t carry through. The checkout flow worked but felt generic.

None of this was catastrophic. The site wasn’t broken. But it also wasn’t selling the shop. If you landed on it without already knowing what letterpress was, you probably didn’t understand why anything on the site cost what it cost. The founder’s actual obsession with the craft, the thing that made the shop worth running, was nowhere in the digital experience.

That’s the real problem with sites like this one. They’re not failures. They’re just quiet. They don’t convert the people who would love the work if they understood it, because the site doesn’t give them a chance to understand it.

The discovery phase

We spent three weeks on discovery. Longer than we normally would for a project this size, but the shop was small enough that there wasn’t much to inherit, and the founder was game to actually dig in.

We did three things.

First, we interviewed her about her craft. Not about the business. Not about her marketing goals. About the actual work. How she chose her presses. Where her paper came from. What she thought the difference was between a good print and a great print. Who her favorite historical letterpress operators were and why. We recorded these conversations. They ran about four hours total.

Second, we interviewed six of her customers. Not with a survey. On the phone. Thirty minutes each. We asked them how they’d found her, why they’d bought what they bought, what they’d given it to or used it for, and what they’d tell a friend if they were recommending her. Those conversations were gold. Not because any one of them said something revolutionary, but because together they told us a consistent story about who was actually buying her work and why.

Third, we spent a day in her studio. Watched her set type. Watched her mix ink. Took probably four hundred photos and about an hour of video. Not as a formal shoot, with lights and a crew. Just a day of being there, documenting what the work actually looks like when it’s happening.

That third thing is what unlocked the project. Everything we did later came from that day.

What we built

The site we ended up with was smaller than the one we replaced. Five main pages instead of twelve. One longer-form essay about the craft, written in the founder’s voice, which the founder had actually never written down before because she’d never thought of herself as a writer. A shop section that was organized by occasion rather than by product category. A journal section that we committed to updating ourselves for the first six months, so it wouldn’t go stale.

The things we didn’t rebuild were as important as the things we did.

We kept the e-commerce platform she was already using, even though a nicer-looking one existed. The one she had worked. She knew how to use it. The friction of switching wasn’t worth whatever marginal visual improvement a different platform would have given us.

We kept most of her existing product photography, with light editing for consistency. We shot new photos only for the hero spots and the about page. Reshooting everything would have doubled the budget without doubling the impact.

We did not build her a membership program. We did not build her a wholesale portal. We did not build her a customer account system. All three had been requested at various points during the project, and all three were things she thought she might want in the future, but none of them would have moved the needle on the actual problem, which was that the current site didn’t communicate what she did.

Phase two can have those features. If the business grows into needing them, we’ll build them then.

The hard conversation in the middle

Around week six, we had the kind of conversation we always end up having on projects like this.

The founder came back with notes on our second-round designs. The notes were small. The hero image should be a little different. The product grid needed more breathing room. The color on one of the headers wasn’t quite right.

But underneath the small notes, we could tell something bigger was happening. We’d put together a design that leaned hard into her actual personality. The copy was punchier than she expected. The photography was more intimate than her previous site. The whole thing felt more like her, which was the point, but which also felt exposing.

We had the conversation. We asked her, straight out, whether she was uncomfortable with the direction because it was wrong, or because it was accurate. She thought about it for a long time. She said accurate.

So we proceeded. We polished the small things. We did not back off from the direction.

That conversation happens on almost every project we do. The client gets what they asked for. They freak out a little. We ask them whether the freak-out is because they got the wrong thing or because they got the right thing and it feels bigger than they expected. The answer is almost always the second. The projects that go badly are the ones where we back off at that moment because the client’s discomfort reads to us like a design problem.

It’s not a design problem. It’s a nerves problem. Nerves pass. Bad design doesn’t.

Launch and after

We launched twelve weeks after kickoff. The founder sent us a note the morning after that said something like “I keep clicking around on it. It feels like my actual shop.” That’s the note you want on launch day.

Numbers over the next six months: traffic up about 40 percent, which is nice but not the headline. Revenue per visitor roughly doubled, which was. Average time on the essay page over three minutes. Customer notes that referenced the journal section started showing up in her inbox.

We’re not going to pretend the website rebuild was the only thing driving those numbers. She was also doing more at the farmers market. She picked up a wholesale account that quarter. Her Instagram was growing. These things don’t happen in a vacuum.

But the site that used to be quiet is now part of the conversation. It converts the people who find it. It tells her story in her voice. It gives her something to point at when someone asks what she does.

What we learned, which we’ve learned on other projects too

Three things that apply to pretty much every small-shop rebuild we do.

One: the real research is not about the business, it’s about the craft. Clients will tell you about their business all day. They’ve been pitching their business for years. What they don’t have well-rehearsed language for is the actual work. The craft. The obsession. That’s where their voice is. That’s what makes their site different from every other site in their category. If you don’t spend time asking about the craft, you’ll end up building a site that could belong to anyone.

Two: smaller is almost always right. Twelve pages became five. Twelve products became organized differently rather than expanded. The site is doing more work with less surface area because every page knows what it’s for. Clients ask for more because more feels safer. Our job is to tell them when less is better.

Three: the hard conversation at week six is the one that matters. Projects like this live or die on whether we’re willing to hold the line on the direction when the client gets nervous. Not stubbornly. Not without listening. But honestly, with the question “is this uncomfortable because it’s wrong, or because it’s accurate?” The answer is almost always the second one. We just have to be willing to ask.

If you’re a small shop reading this and your site feels like the one we rebuilt, you probably know it already. The question isn’t whether. It’s when, and with whom, and whether you’re ready to be seen.