We serve 15 towns. Do we need a page for every town?

The short answer

Fifteen pages that say the same thing aren't fifteen pages of value. Most of the businesses we checked that had town pages had built them from one template. Some changed only the town name. Others had each one reworded, which reads as different and says nothing new.

What we found

The town-name test

Take two of your town pages, blank out the town name in each, and compare them. If you can't tell which is which, the second page isn't telling anyone anything new. We ran that test on every business: we replaced each page's town with the same placeholder and measured how many five-word phrases sibling pages shared.

• On 3 businesses, one each in the UK, the US and Australia, the pages were 98 to 100% identical. Only the town changed.
• On 8, at least half of every page was identical to its siblings.
• The median business was at 25%.

The pages weren't short. The median town page ran to 1,155 words and named its town 14 times. Length and repetition of the town name did the work that local facts should have done.

Rewording isn't the fix

A page rewritten for every town can pass the phrase test and still say the same thing. So we also compared the words each page used. Town pages from two different businesses in the same trade shared a median vocabulary score of 0.44. Sibling town pages on the same site scored a median of 0.90.

11 businesses had town pages that shared under 30% of their phrases but scored 0.80 or more on vocabulary: the same content, reworded. A separate hand check found the other side of this too. One UK cleaning company's city pages shared almost no phrasing at all, yet described the same general service in each. Different wording is not the same as local information.

What the pages offered locally

Local facts were the exception. Only 14 of 42 businesses had any town page showing a postcode that wasn't on their homepage. 19 had a page with its own phone number, which may be call tracking rather than a local office. About a third of pages, 147 of 448, embedded a map.

The hand check showed what useful difference looks like. One UK self-storage company's city pages shared 36 to 42% of their phrasing, but each listed different stores, with their own addresses, for customers to choose between. A plumbing company's two borough pages, by contrast, were 95% identical. The first company's pages answer a different question in each city. The second's don't.

21 of the 78 businesses used a single areas page, or none, instead of a page per town. That's a legitimate choice, not a gap.

What Google says

Google's spam policies describe doorway abuse as pages created to rank for specific, similar searches, and give as an example having pages targeted at specific regions or cities that funnel users to one page. They also name substantially similar pages that are closer to search results than a clearly defined, browsable hierarchy. Its scaled content policy covers making many pages mainly to rank, however they're produced.

None of this sets a percentage, and none of it bans town pages. A business with real branches, or genuinely different services in different areas, has every reason to give each its own page. For Google Business Profile, Google asks service-area businesses to hide their address and keep one profile for the central office with a defined service area, so town pages on the website don't create local offices either.

When a separate town page is worth it

• There's a real branch, team or office there, with its own address and opening hours.
• Services, prices, response times or availability really differ in that area.
• You have local work you can show: jobs, photos, reviews or projects from that town.
• Customers there need to know something specific: permits, property types, access, travel charges.
• The page has its own next step: the right branch, the right phone number, the right booking slot.

If none of these apply, list the towns on one service-areas page, say how far you travel and how quickly, and put the effort into your service pages. If you're merging town pages you already have, redirect each one to the page that replaces it; our guide to redirects covers how.

What AI tools said when we asked

We asked Perplexity and Gemini our question word for word on 24 October. Both said not automatically, and both recommended distinct local evidence and a central areas page. Perplexity drew on 15 sources, mostly search industry publications and agency guides. Gemini cited nothing, recommended starting with three to five towns, a rule of thumb no source supports, and named keyword tools. ChatGPT blocked our logged-out session, so it wasn't tested.

How we checked

We chose businesses that advertise their areas, on purpose. Searches for service-area pages in Yorkshire, Kent, Surrey and Sussex, the Midlands, Texas, Ohio, Arizona, Brisbane, Sydney and Melbourne gave us 80 businesses. For each, we found the largest set of town or suburb pages from its sitemap, its links and its areas page, and checked by hand that they were towns. We sampled up to 12 pages, removed the site's header, footer and menus, replaced the town name, and compared pages within the same service. We also recorded postcodes, phone numbers and maps that appeared on town pages but not the homepage. A separate hand check of three UK businesses on 24 October provided the worked examples.

Limits

• The sample was chosen for having area pages, so it says nothing about how common town pages are across all businesses.
• Text similarity can't see photos, maps or what a business actually does in each town.
• We measured the pages, not how Google treats them. We don't know whether any of these sites has been affected.
• Phone numbers that differ by page may be call tracking, not local offices.