His “Free Estimate” buttons dialed the wrong number. For years.
A painting company with fourteen services, a video library and five-star reviews — and a website that sent callers to a number that wasn't his, named a different company in the copy, and couldn't be properly fixed on the platform it was built on. This is what the rebuild found and what it took.
All County Painters is my brother's company. You'd find that out eventually, so you should know it going in and weigh what follows accordingly.
The site launched recently. There are no traffic or lead numbers to report yet, and I'm not going to invent any. This is a case study about what an audit found and what got built — not about results that haven't happened.
| The client | The situation |
|---|---|
| Who | An owner-operated painting company in Hollywood, Florida, working across three South Florida counties. |
| Already had | Fourteen services, five-star ratings on four directories, a real video library of finished jobs, and years of work behind it. |
| The problem | Call buttons dialing a wrong number, another company's name in the homepage copy, and a locked page builder that made a proper fix impossible in place. |
| What was done | Ten hand-coded pages replacing effectively one, every existing URL preserved, the logo rebuilt from the owner's own concept. |
What the audit found
The business was fine. Fourteen services, a real video library on YouTube, ratings on four directories, and years of work behind it. The website was the problem, and it was worse than it looked.
Several “Free Estimate” buttons dialed a phone number that wasn't his. One digit off. A painting company's most important button, sending every person who tapped it somewhere else.
Nobody knew. That's the thing about a wrong number in a link — it doesn't error, it doesn't warn you, it just quietly costs you the call.
And the estimate email buttons went to an address that had nothing to do with the business either.
Then there was the copy
The homepage named a completely different painting company. Template text, pasted in when the site was built, never taken out. Anyone reading closely would have seen a competitor's name on his interior painting section.
Underneath that: two versions of the business name in use, a social link with a space in it that broke the URL, photo descriptions that described the wrong thing entirely, duplicate headings, duplicate forms, and several service images with no image behind them.
None of this was visible to him. He looked at his site the way owners do — from the top, on a laptop, knowing what it's supposed to say. The things that cost you money are almost never the things you notice.
None of this was a reflection of the work. His Google reviews were already five stars. One customer wrote that he “treats your home as though it is his own,” and another — a drywall restoration client — was impressed enough to hand him more projects.
The business was good. The website was in the way of it.
Why it couldn't be fixed where it was
My first answer was to fix it in place. The wrong phone numbers and email addresses got corrected immediately — those were bleeding.
But the site was built on a locked page builder. A maximum number of sections per page, no site-wide styling, and custom code sandboxed to whichever section you paste it into. You cannot make a consistent site inside that. Every section is its own island.
So I told him the honest version: this can't be fixed properly here, and moving means more work, not less.
He could have said no. Telling a client that the cheap route doesn't exist is a conversation most people avoid, because it sounds like an upsell even when it's true.
The site moved to hand-coded pages on free hosting. No monthly platform fee, no section limits, and every part of the site controlled in one place.
Rebuilding without losing the rankings
The old site had indexed pages, a Google Business Profile, directory listings and a YouTube channel all pointing at it. A rebuild is the most common way a business quietly loses everything it had built up.
The rule was the same as always: change everything about the site, change nothing about how it's found.
| Preserved exactly | Why it mattered |
|---|---|
| Every existing URL | The new pages were built at the addresses that already existed, so nothing needed redirecting and nothing dropped out of the index. |
| The title tag | It was already ranking. One word changed — the business name was inconsistent — and everything else stayed. |
| Name, address, phone | Identical across the site, the Google profile and every directory. Mismatches here are one of the most common local ranking problems. |
| The video library | Real footage of real jobs, embedded on the pages the work matches instead of sitting on one page. |
Ten pages went up where there had been effectively one: a homepage, a services hub, four service pages, an about page, contact, a project gallery and a blog.
The logo: his idea, rebuilt
He had a logo already — a brush with paint strokes. The idea was good: you know what the business does in half a second.
The execution was the problem. A stock brush, only one stroke colored, loose type, the letters “LLC” crammed into the mark, and the whole thing turning to mush at the size Google actually shows it.
I didn't replace his idea. I rebuilt it.
Same brush, redrawn properly. Three strokes instead of four — one for each county he serves. His concept, with a meaning folded into it that wasn't there before.
He was offered a new color direction and chose to keep his blue. That was the right call and it was his to make — the recognition he already had in his market was worth more than a palette I preferred.
The kit that came out of it: horizontal and stacked versions, full color, reversed and single-color black for vinyl and embroidery, a Google profile avatar, and favicons. All of it tested at the sizes that actually break a logo — 96, 48 and 24 pixels.
The word nobody in the market was using
Before writing a page I went through the competition across three counties — the national franchises and the established local companies.
Two gaps stood out.
A specialty finish with real search volume and no local pages
He offers limewash. It's a genuine specialty, it's searched heavily and rising, and not one company in the market had a page about it. He got one.
Every competitor's site was English-only
In South Florida. He's bilingual and so is a large share of his market — and every site he's competing with ignores it. That gap is the same one I write about in Spanish for my own market.
Neither of those is clever. They're just things you find by looking at what everyone else built and noticing what isn't there.
What I told him he didn't need
He didn't need a new logo, and I said so. It wasn't holding the business back, and if the budget had been tight it would have been the first thing to cut. He wanted it, and it was the cheapest moment he'll ever have to do it — a rebrand costs almost nothing during a rebuild and a fortune afterwards.
Two more things I'd rather say than leave implied.
Fixing the phone numbers was probably worth more than everything else combined. Ten pages, a logo kit, a research process — and the highest-value twenty minutes was correcting one digit in a link. That's often how this goes, and it's why the audit comes before the design.
In a different rebuild, for a clinic in North Carolina, the client reported bookings for a service she had offered for years and never advertised properly. Same principle, opposite direction: there, the words were missing. Here, the wrong ones were present.
And a new website doesn't make a business good. His reviews were already five stars, the video library already existed, the work was already the work. A rebuild removes what's in the way. It doesn't create what isn't there.
Questions people ask about this
How do I check whether my own website has a wrong phone number?
Open your site on a phone and tap every button that is supposed to call you, one at a time, including the ones in the header, the footer and inside each service section. Do not read the number, tap it, and look at what the dialer shows before you hang up. The number displayed on the page and the number inside the link are two different things, and they can disagree without anything looking wrong.
Can a website be rebuilt without losing search rankings?
Yes, if the addresses are kept. The most common way a rebuild destroys traffic is that pages move to new URLs with no redirects, so everything Google knew about them stops existing. Building the new pages at the addresses that already exist avoids the problem entirely. Titles, headings and contact details being preserved matter for the same reason.
Why move off a website builder at all?
Sometimes there is no reason to. Locked builders are limiting when a site needs consistent styling across many pages, or custom code that works site-wide, or more sections than the platform allows. If a business needs a handful of simple pages and never touches the design, staying put is usually the cheaper answer. The honest test is whether the platform is preventing something the business actually needs.
Does changing a logo hurt search rankings?
No. Search engines do not read a logo image as a ranking signal. What matters is that the business name is spelled consistently on the website, the Google Business Profile and every directory listing. If a rebrand changes the name itself, that consistency work is the real task, and it is worth doing carefully. Changing only the artwork has no search impact at all.
Should a website be in Spanish as well as English?
It depends on the market. In parts of South Florida and the Treasure Coast a large share of homeowners search in Spanish, and most local competitors publish in English only, which leaves those searches uncontested. The important part is that a Spanish version has to be genuinely written rather than machine-translated, because a page that reads as translated does the opposite of building trust.
Is this case study biased because he's your brother?
Probably, and that is why it is stated at the top rather than buried. The verifiable parts are the ones worth judging: the wrong phone number in the call links, another company's name in the homepage copy, and the platform limits that made a proper fix impossible in place. Those were either true or they were not. The parts that involve my judgment should be weighed knowing the relationship.
When did you last tap your own call button? Actually tap it.
Send me your business name and I'll check the links, the listing and the spelling. Free, and I'll tell you if there's nothing wrong.
