SEO for multiple locations: what scales, and the part that does not
You are not ranking in several cities, you are running separate local contests. Distance is a documented factor and it does not scale with effort.
Guides on this query agree on the mechanics, and they are right about most of it: a page per location, subfolders rather than subdomains, unique body copy, consistent name and phone, local links. What none of them says out loud is the constraint that decides how far any of it gets you. You are not ranking in several cities. You are competing in several separate local contests, and one of the three factors Google documents is distance, which does not scale with effort. Knowing that before you build forty pages changes which cities go on the list.
Quick answer
- Each location is its own contest. There is no single ranking across cities.
- Distance is a documented factor and it cannot be optimized. It caps what any page can do.
- A location without an address competes differently, and mostly not in the map.
- The page-per-city approach works, and only when each page says something true about that city.
| What scales | What does not |
|---|---|
| Structure, templates, technical setup | The address you do not have |
| Consistent name, address, phone everywhere | Proximity to the person searching |
| A process for collecting reviews per branch | Prominence, which accrues per location and slowly |
| Local links, one market at a time | Ranking in a city where you have no presence |
What kind of problem this is
The mental model that produces forty thin pages is treating this as a content problem: more cities, more pages, more coverage.
It is a presence problem wearing a content problem's clothes. In a query with a map, what decides the result is the combination of relevance, distance and prominence, and only two of those respond to work. A page for a city where you have no address can rank for the non-map queries in that market, and it will not put you in the map pack there. Those are different results on the same screen, and conflating them is where the budget goes.
So the useful first step is not building pages. It is sorting your target cities into two lists: where you have a real presence and can compete for everything, and where you can only compete for the queries that do not carry a map.
How do you rank in more than one city?
By treating each location as a separate contest. Google documents three local factors and distance is one of them, so a branch competes mainly in its own radius. What scales is structure, consistency and process; what does not scale is proximity, and no page fixes an address you do not have.
The structure, which is the easy half
Subfolders, not subdomains. For a business with several branches, yourdomain.com/locations/city keeps everything on one domain and one analytics property. Subdomains split what you have built and give you several small sites instead of one that adds up.
One page per location, and each one verified. The page and the listing are different assets that reinforce each other: the listing decides the map, the page decides the rest of the screen, and the field that most often goes wrong on the listing is the primary category: optimizing your Google Business Profile.
Name, address and phone identical everywhere. It is dull and it is one of the highest-return items on this list, because contradictions between directories are noise the system resolves at your expense. The detail is in local SEO for small businesses.
And the reviews collected per branch, not centrally. Reviews attach to a listing, not to a company, so a chain with four hundred reviews on headquarters and none on the branches has four branches with no prominence at all: how to get Google reviews.
The hard half, which nobody prices
Distance caps each page before you write it. A branch competes consistently within a short radius and intermittently in a ring around it. Outside that, rarely, no matter how good the page is. That is not a failure of the page; it is the factor that does not move, and it is the one thing the guides on this topic never price in: how to rank in Google Maps covers what the three factors actually do.
Prominence accrues per location and it accrues slowly. A branch opened last month starts from zero even if the brand is twenty years old. There is no way to transfer it, and the honest projection for a new location is months, not weeks.
And a city where you have no address is not a location, it is a market. You can compete there for queries that do not trigger a map, with pages that say something true and specific about that market. You cannot appear in its map pack consistently, and any provider promising otherwise is either selling a listing at an address you do not occupy, which violates Google's guidelines, or describing something else.
The test that separates a location page from a template
Every guide warns against cloning pages and swapping the city name. None gives a way to check it, and there is one that takes ten seconds:
Cover the city name and read the page. If it still makes sense for any other city, you do not have a location page. You have a template with a variable, which is exactly the pattern Google's guidelines describe when they talk about pages made to capture queries without adding value. The Spanish guide on páginas por ciudad goes into where that line sits and why the usual advice about a percentage of unique text misses it.
What passes that test is specific and unglamorous: the actual service area and drive times, the work done in that market, what changes in that municipality's rules, the team that works there. It does not need to be long. It needs to be true and impossible to paste into the next page.
When not to build these pages
When the branches are not verified. None of this applies until each listing appears in the map, and that has its own list of causes, most of them about data rather than ranking: why your business does not appear on Google Maps.
When there are two locations in the same city. They compete with each other rather than adding coverage, and the fix is usually one strong page and two listings.
And when the business is service-area rather than premises-based. That case has its own rules and its own listing type, and copying the multi-branch playbook into it produces pages that never rank.
Mistakes that repeat
- Building pages for cities with no presence. They compete for the non-map queries only.
- Treating it as one ranking. There is no cross-city position; there are separate contests.
- Collecting reviews centrally. Reviews attach to a listing, so branches start at zero.
- Swapping the city name and calling it unique. Cover the name and read it: if it fits anywhere, it is a template.
- Using subdomains. They split what you have built into several small sites.
- Promising a timeline for a new branch. Prominence accrues per location and it takes months.
Data and transparency
That Google documents three factors for local results, relevance, distance and prominence, comes from its own guidance, checked on 13 August 2026. That reviews attach to an individual listing rather than to a business as a whole is documented behaviour of Google Business Profile, checked the same day. That the first page of results for this query covers structure, unique content, consistency and local links but does not mention distance among the constraints is an observation from the same day, recorded as an omission rather than as a source. No radius, number of reviews or timeline is given here: all three depend on competitive density in each market and on the category, and any general figure would be invented. The rule of covering the city name to test a location page is operating judgement from work on a portfolio recording more than 300 million annual impressions in Search Console. Verified as of August 2026.
Primary sources, opened on 13 August 2026: Google's three local ranking factors.
What this changes
There is a version of this project that goes well and a version that produces forty pages nobody reads, and the difference is decided before any of them is written.
The guides describe the machinery, and the machinery is real: structure, consistency, reviews, links. What they leave out is that the machinery has a ceiling per city, set by where you actually are. A business that accepts that ceiling builds four excellent pages for four markets it can win and treats the fifth city as a long game. A business that does not accept it builds forty, spends a year, and concludes that local SEO does not work.
The question worth asking before the first page is not how to rank in more cities. It is which cities are actually mine, and which am I renting attention in? That answer costs an afternoon of searching from different points, and it is the cheapest part of the whole project.
Frequently asked questions
Can I rank in a city where I have no office?
For queries that do not trigger a map, yes, with a page that says something true and specific about that market. For the map pack itself, not consistently, because distance is one of the three factors Google documents and you have none of it there. Any provider offering a listing at an address you do not occupy is proposing something that violates Google's guidelines.
Should each location have its own page or its own website?
Its own page, on the same domain, in a subfolder. Separate websites split whatever the brand has built into several small properties that each start over, and they multiply the maintenance. The exception is a business whose locations are genuinely different brands, which is a different problem.
How many location pages is too many?
The number stops being the question once each page fails the test: cover the city name and see whether the text still fits anywhere else. Four genuine pages beat forty templated ones, and forty templated ones can attract the kind of attention nobody wants. Build them at the pace at which you have something real to say about each market.
Do reviews from one branch help the others?
Not directly. Reviews attach to the individual listing, so a new branch begins with none regardless of what headquarters has accumulated. That is why the collection process has to run per location from the day it opens, and why the honest projection for a new branch is months rather than weeks.
Is a service-area business handled the same way?
No, and copying this playbook into it is a common and expensive mistake. A business that travels to customers uses a different listing configuration and competes on a different surface, which means most of the page-per-city advice above does not apply and some of it is actively counterproductive.
Most sites do not have a ranking problem
They have a what-happens-next problem. You can rank first and still sell nothing. The diagnostic looks at both and tells you which one is costing you money.
See the diagnostic