The problem
This is the second half of the Shuttersmith story. Our SEO/GEO work on the client’s first business worked well enough that they decided to launch a second, complementary one: plantation shutters at trade price, with a self-install option for customers who’d rather measure and fit their own than pay for a full install.
That is a genuinely different thing to sell online. A shutters install is a consultation — someone visits, measures, quotes. A trade-price self-install kit is closer to a product: the customer needs to arrive at a price themselves, trust that the measurements they took are the ones that will be cut, and understand exactly what they are and aren’t getting for the lower number. A brochure site with a contact form doesn’t do that job. It needed real quoting tools.
What we built
A full Next.js site built around a measure-it-yourself flow. Customers size their own windows using an on-site calculator, then choose between a professional install and a DIY self-install kit priced roughly 20% below the installed rate. Pricing is transparent and flat — the client’s call, and one we built the calculator around rather than around a sales team.
Flat pricing is worth dwelling on, because it changes what the site has to do. When the price is negotiated, the site’s only job is to generate a lead and hand it to a person. When the price is published and fixed, the site has to answer the whole question — sizing, options, what self-install actually involves, what it costs — well enough that the customer can decide without a call. That is more work to build and considerably better to own: the qualifying happens before the enquiry rather than after it.
The product range is deliberately narrow: one frame material, one louvre size. That is not a limitation we’re working around — it is the decision that makes the DIY tier possible at all. Every extra option in a made-to-order product is another way for a self-measured window to come back wrong, and on a kit the customer cuts nothing and returns nothing, so the measurement has to be right the first time. Narrowing the range narrows what can go wrong with it. Pricing itself is never computed in the browser: postcode and dimensions go to the server, which is the only thing allowed to decide a price, and both tiers are priced together so switching between “measured & installed” and “do it yourself” is an instant comparison rather than a second wait. Postcode also decides which tiers are even on offer — installation isn’t viable everywhere delivery is, so the site checks coverage before it lets someone quote a tier it can’t actually deliver.
Launching with SEO/GEO already in place
SEO/GEO was built in from the start, the same way it was for Shuttersmith and Edwards Kirby: full metadata, sitemap.xml, robots.txt, an llms.txt for AI-search visibility, and structured data throughout.
Doing it at launch rather than later is the cheapest this work ever gets. A new site has no legacy URLs to preserve, no accumulated content written without schema in mind, and no existing rankings to protect while things move — every argument for deferring it disappears. It also means the business isn’t starting from zero on search the way most launches do, and that when an AI assistant is asked about trade-price shutters, there is something legible for it to read on day one instead of six months in.
There is a second reason we’re paying attention to this one. Same owner, same product category, same market — but one site was retrofitted and the other launched clean. Over the next few months that gives us an unusually direct comparison of what building it in is worth against bolting it on, on data we own rather than on a case study someone else published.
Why we built it this way
Shuttersmith’s audit found a fault that had nothing to do with the site’s own code: the hosting it ran on was silently blocking several AI crawlers at the server level, invisible until we tested it directly. That’s the kind of thing a retrofit finds after the fact and a clean build can rule out before launch — so before this site goes live, the same direct crawler test runs against it first, not after something looks wrong in the analytics.
Analytics also goes in at launch rather than after, for the same reason it mattered on Shuttersmith: without it, there's no way to know what the first month of a brand-new site actually did. That matters more here than usual, because it sets up a comparison nobody else can publish — the first month of a launched-clean site against the first month of a retrofitted one, same owner, same market, side by side.
Where we are
The build is complete and in staging, finishing final pricing sign-off ahead of going live. This page will carry real launch-month numbers once there are any to report — not before.