A landscaping company covers Phoenix, Scottsdale, and Tempe. A home inspector covers half the East Valley. A branding studio takes clients from Flagstaff to Tucson. All three face the same question: does each city get its own page, or does the whole service area live on one?
Most businesses answer it badly. They spin up a page per city, swap the name in a paragraph that is otherwise identical, and call it local SEO. That pattern has a name, doorway page, and it has been a liability for a long time. Answer engines make the liability worse.
The doorway page problem, briefly
A doorway page exists to rank for a location, not to serve a reader. City name in the H1, city name in the URL, and text underneath that reads identically to the page for the next city over. Google has penalized the pattern for over a decade because it adds volume to the index without adding information.
The instinct behind it is not wrong. Businesses that serve multiple cities should be found for each one. The execution is what fails, because swapping a proper noun is not the same work as actually describing what changes city to city.
Why answer engines are even less forgiving
Classic search could occasionally let a thin city page rank on the strength of the domain around it. An answer engine does not have that mercy. It reads the page, looks for something specific enough to state as fact, and moves on if it finds nothing.
A templated paragraph gives it nothing. No neighborhood names, no distinguishing detail, no proof tied to that location. The page does not rank poorly, it simply never gets pulled into an answer, because there is nothing on it worth quoting. We wrote about this same mechanism in our post on why AI recommends your competitor: engines favor whoever has evidence beyond a claim, and a duplicated city page is a claim with no evidence attached.
What a real service area page needs
Start with what actually differs. Permits and codes that vary by municipality. Neighborhoods or landmarks a local reader would recognize. Response times or crew availability specific to that area. Jobs completed there, named if the client agrees, with real detail instead of a generic mention.
Then build the FAQ section from real questions, not invented ones. If a Scottsdale homeowner asks something different from a Mesa homeowner, that difference belongs on the page. The phrasing buyers actually use is the material worth putting there, and it rarely repeats identically from city to city.
Add Service schema with an areaServed property naming the specific city, not just the metro. It is a small structural signal, and it costs nothing to include once the page has real content to attach it to.
The architecture question: one page or many
We made this argument about services in an earlier post, and it applies to locations the same way. One Service, One Page argued against bundling distinct services onto a single page because it gives both search engines and answer engines nothing specific to match a query against. Cities work the same way in reverse. If a city has enough real substance to fill a page, give it one. If three neighboring cities genuinely get identical service with no local difference worth stating, one honest page naming all three beats three thin pages pretending otherwise.
The wrong move is deciding this by ambition instead of by substance. A twelve-page location grid because a competitor has one is exactly the doorway pattern that gets ignored by machines and, eventually, penalized by Google.
Corroboration has to show up at the city level too
A service area page reads as more credible when the rest of the record agrees with it. Reviews that mention the city by name. A Google Business Profile with the correct service area set. Local directory listings that match the same name, category, and cities served everywhere they appear.
This is entity consistency applied at a smaller scale. We wrote before about how scattered agreement across surfaces reads as trust and disagreement reads as noise. A city page is one more surface that either lines up with the rest of your record or quietly contradicts it.
Build fewer pages, and mean each one
A multi-city business does not need a page for every zip code in the metro. It needs pages for the cities where there is real volume and real proof, each one carrying something a competitor's templated page cannot. Everything else can live on a shorter list within a single service area page, described honestly instead of inflated for the index.
Deciding which cities earn a dedicated page, building the local schema and proof each one needs, and keeping the whole record consistent across Phoenix, Scottsdale, and every other market a business serves is part of the work inside our SEO / AEO / GEO Foundation, from $7,500. A city page only earns a citation when it says something a copy-paste page cannot.