Google just told us how to name a business if you want to show up in search and AI results
This one's been bugging me for years, and Google's John Mueller finally said the quiet part out loud. If your business name is a string of keywords, Google won't treat searches for it as searches for your brand. And AI search? It works the same way. If you're running a home service company in Tampa and an SEO told you to register a DBA like "Best Tampa Roofing Pros," you're about to read why that advice is actively hurting you.
What Mueller actually said
Someone hopped on Bluesky and complained that their site doesn't show up when they search for their own site name. They wanted Google to fix it. Mueller's answer was, honestly, a little blunt:
If your "site name" is made up of generic, competitive keywords, Google does not assume people are looking for you specifically when they search for those words.
That's the whole game right there. Google's job is to figure out intent. When someone types "roofing company," Google has zero reason to assume they meant your roofing company called "Roofing Company." It assumes they meant the category. So you get buried under the entire industry, and the brand searches you think you're earning aren't being credited to you at all.
And here's the part that makes this a bigger deal than a normal Mueller quote. The same logic now runs the AI engines. ChatGPT, Gemini, Perplexity, Claude, Google's AI Overviews. They all rank entities, not URLs. An entity is a thing the model recognizes as a real, distinct brand. A generic keyword string isn't an entity. It's a category. Models don't cite categories. They cite brands. That's the whole game behind AEO (answer engine optimization).
Why this hits home service companies the hardest
Roofers, plumbers, HVAC, garage door, electricians, landscapers. You guys get this advice from cheap SEO companies more than anyone else. The pitch sounds reasonable on day one. "Register a DBA called Dallas Roofing Pros, buy dallasroofingpros.com, and you'll rank for everyone searching dallas roofing." It worked a little bit in 2012. It hasn't worked for a long time, and in 2026 it's a slow-bleed problem you might not even notice until you're a year in.
Here's what actually happens. A homeowner in Plano hears about you from a neighbor. They Google your DBA, "Dallas Roofing Pros." Google sees a generic category query, serves up the local pack with five competitors, and you might not even be in the top three. The homeowner picks someone else. You never knew the lead existed. Now multiply that by every word-of-mouth referral, every yard sign drive-by, every truck wrap someone glances at on the freeway. Fixing this is half of what real local SEO work is in 2026.
Why keyword DBAs are SEO poison in 2026
- You can't build branded search volume. Branded search (people typing your name into Google) is one of the strongest trust signals modern SEO has. If your name is keywords, every search looks like a category search, not a brand search. You get zero credit.
- AI engines won't cite you. LLMs need a clear entity to point at. They lean on Knowledge Graph, Wikidata, third-party press, Reddit threads, GBP. None of that gets attached to "Best Houston Plumbing." It gets attached to real names like Roto-Rooter or ARS Rescue Rooter.
- You can't trademark a category. The USPTO won't trademark generic descriptive phrases. So you have no legal protection. A competitor down the street can call themselves Best Houston Plumbing too, and you can't stop them.
- You're invisible offline. Try saying "thebestdfwroofingexperts.com" on a radio ad. Try fitting it on a hat. Try getting a referral when someone has to stop and ask, "wait, what's the name again?"
- Google's Knowledge Panel won't trigger. No Knowledge Panel, no entity authority, no AI citations. The whole chain breaks at step one.
- Yelp, Angi, BBB, and citation sites get confused. Multiple businesses with similar generic names create NAP soup. Google can't tell who's who. Your Google Business Profile can even get suspended for the duplication, and your local pack ranking takes the hit either way.
Branding isn't a luxury for home service companies. It's the moat.
The home service industry has a branding problem and most owners don't realize it. You're competing in a category where customers can't tell most companies apart. Same trucks, same uniforms, same Google ads, same "family owned since whenever" tagline. The only thing that separates a five-year company from a one-year company in the customer's head is whether the name sticks.
Think about the home service brands that broke through. Mr. Rooter. Roto-Rooter. Mike Diamond ("the smell good plumber"). One Hour Heating and Air. Benjamin Franklin Plumbing ("the punctual plumber"). Chuck's Roofing. None of those are descriptive keywords. They're names. They're characters, mascots, promises, or just plain weird enough to remember. And every one of them earns branded search volume that the keyword-DBA crowd never sees.
A study from the Local Search Association a few years back found that branded queries convert at multiples of generic queries, because the searcher already trusts the name. That's the whole point. A real brand collapses the customer's decision down to one step. A keyword DBA puts you back in a five-way comparison shootout every single time.
What a real home service brand looks like
You don't have to be clever. You don't need a marketing agency to make this up. Some of the strongest local brands I've worked with are just a last name plus the trade, or a single short word that the owner liked. The criteria are simple:
- Sayable in one breath. Two to three syllables for the brand word. If a customer has to ask you to repeat it on the phone, it's already wrong.
- Trademarkable. Something unique enough that the USPTO will register it. If it's a description, it isn't a brand.
- A real character or hook. A mascot, a color, a promise, a tagline, a uniform. Mike Diamond's whole brand is "the smell good plumber." That's a hook a customer remembers six months after the truck drove by.
- Owns the .com. Not the .net, not the hyphenated version. Even a slight variation like getheritageplumbing.com beats heritageplumbingnow.com if the clean version isn't available. (And once you own it, build the site right. Cheap template sites bleed conversions. A real conversion-focused build pays for itself in a quarter.)
- Consistent everywhere. Same exact name on Google Business Profile, Yelp, Angi, BBB, Facebook, your website, your trucks, your invoices. NAP consistency isn't a tip, it's table stakes.
- Visual identity that doesn't change. Logo, color, font, truck wrap. Pick it once, stick with it for ten years. That's how a brand burns into a neighborhood's memory.
"But my keyword DBA is already ranking..."
I hear this every couple of weeks. Usually from someone who's been in business 8 to 15 years and built real authority on a keyword domain. Fair point. Don't panic-rebrand if you've got that kind of history. The migration cost can outweigh the upside if you're already getting calls.
But here's what I'd do. Start treating the keyword name as a legacy URL. Build a real brand name on top of it, even if the domain stays the same for now. Get the brand into the Knowledge Graph. Earn press under the brand name. Run paid ads under the brand. Plan an eventual migration once the entity signals are strong. Most clients I've helped do this over 12 to 18 months and don't lose ranking, because the entity bridge is built before the URL changes.
If you're under two years in business, or you haven't built much authority yet, the math almost always favors moving now. The longer you wait, the more rebuild you're signing up for later.
A warning about SEO companies still pushing keyword DBAs
If your current SEO told you to file a DBA like "Best City Service Pros" and buy the matching domain, that advice is straight out of 2012. They're optimizing for a Google that hasn't existed in a decade, and they're definitely not paying attention to how AI search picks brands to cite. You're going to spend a year paying for a strategy that gets quieter every quarter as more search shifts to LLMs. That's not opinion. That's where the data is pointing.
Where this leaves you
Mueller didn't drop a bomb here. He just confirmed what's been true for a few years now and added the AI search piece on top. Google needs a real brand name to attach search intent to. AI engines need a real entity to cite. Home service customers need a name they can actually remember. All three of those things point to the same answer, and it isn't a DBA full of keywords.
Pick a real name. Build a real brand. Earn branded searches. That's the playbook for the next decade of local SEO, and it's the only one that survives both Google's next core update and the slow takeover by AI Overviews and chat search. Stuart McHenry Consulting has been helping home service companies make this exact call for years now, and I can tell you the businesses that committed to a real brand five years ago are the ones running away with their markets today. If you want a second opinion on yours, the contact page takes about 30 seconds.
Send me your current name and I'll give you a straight answer on whether to keep it, tweak it, or rebrand. No pitch, no upsell, just a real take from someone who's seen this exact problem play out a hundred times.
Get in touch