He Asked for a Proper Locksmith Website. He Got Mission Control.

Sean Hamilton's brief for Lockerfella was beautifully simple.

He wanted the site to look proper. Not cheap. Not rushed. Not a national chain pretending to be local. He wanted people to land on it and think, "Right, this bloke looks genuine. He knows what he's doing."

That was it.

He didn't ask for schema. He didn't ask for AI Discovery Files. He didn't ask for E-E-A-T signalling, entity clarity, technical SEO, or 18 carefully written local area pages. As Mark McNeece put it during a post-launch chat, "If you had, I'd have checked whether you'd been replaced by a marketing agency."

But that simple brief produced one of the strongest small-business builds we've shipped in a long time. A perfect 100/100/100/100 mobile PageSpeed score on day one. Strong rankings in Google and AI search inside ten days. And, crucially, a website that actually looks and reads like the man behind the business.

This is the story of how that happened, what the brief actually meant once we took it seriously, and what other small business owners can take away from it when it's their turn to brief a designer.

Prefer the unedited primary source? Read the full Sean Hamilton interview on which this case study is based.

"One Man, One Van, One Phone Number"

That phrase is the whole brief, really.

Sean said it casually about ten minutes into our first call, and it ended up doing more strategic work than anything else he told us.

It tells customers there's no call centre between them and the locksmith. No mystery subcontractor turning up at midnight. No dispatch system passing the job to whoever's free. If you ring Lockerfella, you get Sean. If you answer the door to a Lockerfella van, the man stepping out of it is the one who quoted you on the phone.

That isn't branding. That's trust. And for a locksmith, trust is the entire deal.

Sean Hamilton didn't think of it as marketing. He just thought of it as the truth.

"If you ring me, you get me. If I say I'm coming, I'm coming. If I give you a price, I stick to it. I've never liked all that messing about where someone rings a locksmith, gets told one thing, then the price changes when the van turns up."

Sean Hamilton, Lockerfella

That sentence runs through every page on the site, even though it appears in full on none of them. It's the spine.

The Trust Problem in the Locksmith Trade

Here's a fact that doesn't get said often enough in web design conversations: the locksmith trade has a public trust issue, and the decent locksmiths really hate it.

People worry they'll be pushed into work they don't need. Charged a call-out fee they didn't understand. Told a lock has to be drilled when it doesn't. Quoted one price on the phone and then handed a different one when the van pulls up.

Sean is one of those people who finds that flat-out depressing. Not because it makes him look bad by association (it does), but because of the kind of customer who tends to ring a locksmith.

"If you're locked out or your door won't secure, you're vulnerable. You're stressed. You might have kids with you. It might be late. You need help, and that's when some people take advantage. I think that's unforgivable."

Sean Hamilton, Lockerfella

When a client tells you that with that level of feeling, your job as a designer changes. The site doesn't just need to look professional. It needs to obviously not be one of those sites.

Trust didn't end up as a "values" section near the footer. It ended up baked into the way every page is structured.

What Trust Looks Like on a Page

On Lockerfella, trust is not something we said. It's something we showed.

• No call-out fee, ever.
• All-in price quoted on the phone before he leaves the van.
• Itemised receipt for every job, parts and labour separated.
• Non-destructive entry attempted first wherever the lock allows.
• One person, start to finish.

That list lives near the top of the About Sean page, not buried in a values section nobody scrolls to. It's stated plainly. Then it's reinforced everywhere else: in the service pages, in the prices, in the area pages. Because if you say one thing on the homepage and contradict it three clicks deep, customers notice.

Service pages this clear are also exactly the kind of evidence Google's March 2026 core update looked for: real, specific, first-hand. Helpful in a way only the actual business could write.

Then the Prices Page Does the Same Job, Only Louder

Most locksmith websites bury pricing or refuse to publish it at all. "Call for a quote." "Prices vary." "Get in touch for tailored pricing."

Lockerfella publishes the prices.

Look at what that does for someone who's locked out at 9pm with two cold kids and a dead phone:

1. They know roughly what it'll cost before they pick up.
2. They know there's no surprise call-out fee.
3. They can rule out cowboys whose answer to "how much?" is "we'll see when we get there."

That's a stronger trust signal than any "Trusted by 5,000 customers" banner you can put on a homepage. It's also the kind of clarity that helps the page get cited by AI assistants when someone asks ChatGPT or Claude how much a locksmith costs in the West Midlands.

"I Didn't Realise I Was Getting the Locksmith Version of Mission Control"

About halfway through the post-launch chat, Sean said something I'm definitely keeping.

"I thought I was getting a smart website. I didn't realise I was getting the locksmith version of mission control."

Sean Hamilton, Lockerfella

That's the bit that matters for anyone reading this who's about to brief a web designer for their own business.

You probably won't ask for the technical layers. You shouldn't have to. You shouldn't need to know what schema is, what an AI Discovery File is, why one area page should be different from another, or why Google might trust one page more than another.

Your job is to bring the business truth. The designer's job is to turn that truth into the right website, with the right structure, talking to humans, search engines, and AI systems all at once.

The client should bring the business truth. The designer should bring the structure.

What "the Right Structure" Actually Meant

Sean is based in Brewood. He covers a real, geographically coherent catchment running outward through Codsall, Penkridge, Wolverhampton, Stafford, Cannock, Dudley, Walsall, West Bromwich, Birmingham and Sutton Coldfield. Eighteen towns in total.

The lazy version of this is well known. One "Locksmith in [Town]" template with the place name swapped in. "We are fast and reliable in [Town]. Call us today." Then 18 of those.

Google has seen enough of those pages to fill a skip. So have customers.

The proper version is harder. Each area page on Lockerfella talks about what you actually find when you turn up there.

The Codsall page mentions the kind of housing you'll see around Histons Hill, Watery Lane and Birches Road, and explains why anti-snap cylinder upgrades are common in that mix. The Wolverhampton page has a different feel, because Wolverhampton has a different mix: older terraces, newer estates, uPVC-heavy developments and the multi-point gearbox failures that come with them. Stafford is Victorian and Edwardian terraces alongside newer estates around Baswich, Wildwood and Weeping Cross.

Cannock is post-war and later housing, mostly euro cylinders and multi-point uPVC, with pockets of older properties where five-lever mortice locks still appear. The Cannock locksmith page walks through a typical job in that area: doorstep arrival, secure first, then the calm conversation about whether the lock can be reused or needs replacing.

That kind of content isn't glamorous. But it does the job. It shows local understanding. It helps customers feel seen. It helps search engines understand relevance. And it gives AI systems far more to work with than "we provide fast locksmith services in your area."

A Brief Detour on Birmingham

I'm told, by a man who actually covers Birmingham, that the city has its moments.

The point underneath the joke is serious enough. Some areas turn up jobs that aren't in the standard pricing brochure, and good local content acknowledges that without dramatising it. The Birmingham area page covers what's actually needed: response time variance depending on which side of the city, the kinds of doors and locks common to specific postcodes, and the sort of after-incident callouts that come up.

That is what separates a useful local page from a doorway page Google has been deindexing for a decade.

Specific Content Has Fingerprints On It

A generic locksmith page says: "We offer fast and reliable locksmith services."

A Lockerfella page says: Sean is based in Brewood, answers the phone himself, charges no call-out fee, works with five-lever mortice locks, anti-snap euro cylinders and uPVC multi-point gearboxes, and understands the property types in the towns and villages he covers.

The second one has fingerprints on it. It belongs to one specific business. It can't be copied to another locksmith site without the new owner sounding like they've stolen it. That's exactly what good content should feel like.

Good SEO should not feel like SEO. It should feel like the page is useful. The Codsall page is for people in Codsall who want to ring a locksmith. The Cannock page is for people in Cannock. The Wolverhampton page is for people in Wolverhampton. They each get content that fits their town.

Search engines and AI systems work that out as a side effect.

E-E-A-T Isn't a Label. It's Evidence.

People sometimes talk about Google's E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness) as if it's a tickbox you stick on a page and Google sends a fruit basket.

It isn't.

E-E-A-T is evidence. Experience means showing you've actually done the work. Expertise means writing about it the way only someone who's done it can. Authoritativeness means the rest of the site (and the wider signals around it) back up what one page says. Trust means people can see who you are, where you work, and how you operate.

Google's Search Quality Rater Guidelines call out "first-hand expertise and a depth of knowledge" as the thing that lifts a page above the generic crowd. Real stories. Real callouts. Real numbers. Real opinions. Things the AI training data didn't already have.

Lockerfella checks all four because the brief was honest in the first place. Sean is real. He's Brewood-based. He's certified, DBS-checked, insured. He offers a 12-month workmanship guarantee. He doesn't charge a call-out fee. The site says so plainly, the prices back it up, the area pages reinforce it, and the way the content is written tells you a working locksmith reviewed every word.

That is what E-E-A-T looks like in practice on a small service business website. Not a "Trust Centre" page. Not a banner. Real proof, scattered through the site as a side effect of the business being properly described. We've covered the same point in detail in our look at why clear, structured writing is being dismissed as AI, and why that's the wrong instinct: clarity isn't a fingerprint, it's a feature.

The Technical Layer Sean Had No Idea Was There

The visible layer of Lockerfella is the design, the photography, the area pages, the pricing. Underneath, the site also ships with a full set of AI Discovery Files at launch: llms.txt, ai.txt, ai.json, identity.json, brand.txt, faq-ai.txt, developer-ai.txt and robots-ai.txt.

That gives ChatGPT, Claude, Gemini, Perplexity and the AI overview layers in Google a clean, machine-readable description of who Lockerfella is, how the brand should be referred to, what the key facts are, and what AI systems are permitted to do with the content.

When you publish those properly on day one, you skip the bit where AI systems guess. We've shown what a properly written one looks like in our piece on why so many llms.txt files fail: most are auto-generated URL dumps with no business identity in them. Lockerfella's are not.

The technical case study side gets covered in detail elsewhere:

• The 365i Lockerfella PageSpeed case study explains the 100/100/100/100 score, the lean build, and the hosting performance.
• The Pressforge Lockerfella AI search case study explains how the site ended up cited in AI answers within ten days of launch.

Lean build, no plugin soup, no animation library doing cartwheels in the background because someone discovered JavaScript in 2009. Hosting sets the floor. The build sets the ceiling.

The Brief vs the Build

The most useful way to summarise Lockerfella is to put what was asked next to what was delivered. Reading them in the same room is the bit that surprises most business owners.

The brief Sean gave us, and the build that came out the other side.

What Sean asked for	What the brief actually meant in practice
"Make it look proper"	Distinctive design, working van photography, mobile-first hero, and zero stock-photo locksmith clichés
"Sound like me, not a call centre"	First-person tone throughout, with the trust commitments written the way Sean talks
"One man, one van, one phone number"	About page evidence, single phone number across the site, no fake "team of specialists" copy
"Make sure people can find me locally"	Eighteen unique area pages with property mix, postcodes, response time, and recent jobs
"I don't want it to be slow"	100/100/100/100 mobile PageSpeed at launch, lean build, 365i hosting, no plugin soup
(Not asked for, ever)	BlogPosting, LocalBusiness and Service schema, FAQPage schema, breadcrumbs, all 8 AI Discovery Files at launch

Recent Jobs as Evidence

One of the small ideas that did the biggest work: each area page has a "Recent Jobs" section where Sean writes up a few real callouts in his own voice. Anonymised, dated, with the kind of detail that proves the locksmith was actually there.

This is content that nobody else can write. A national locksmith chain can't fake a 1990s uPVC door in Heath Hayes that has dropped a quarter of an inch and binds on the keep. Either you have done that job in that town, or you can't write the entry. Customers can feel the difference. So can AI systems trying to figure out which locksmith pages are written by people who actually do the job.

It's the same principle as the case studies on our own trades portfolio page: specific, dated, unfakeable.

What This Says to Other Business Owners

If you're about to brief a designer for your own business, you don't need a glossary of SEO terms. You need to be honest about how the business actually works.

What do you do? Where are you based? Who do you do it for? How do you charge? What do you refuse to do, and why? What kinds of customers tend to find you, and what state are they usually in when they ring? What separates the work you're proud of from the work you'd rather not talk about?

Those answers are the brief. They're also, oddly, the best raw material for everything else: schema, AI Discovery Files, area pages, trust signals, the lot. None of which you should have to ask for by name.

"In 2026, Google is far more focused on who is behind a website, whether that business has real-world credibility, and whether its content can actually be trusted."

Code Guys, E-E-A-T and trust signals 2026

The Lockerfella build only worked because Sean told the truth about how he runs his business. Once we had that, every other layer followed naturally. The schema described a real local business. The AI Discovery Files described a real owner. The area pages described real towns and real housing. The pricing described real money. The trust signals described real commitments.

None of it had to be invented. None of it had to be exaggerated. None of it had to read like marketing.

That's the difference between a website that exists and a website that works.

Frequently Asked Questions