Forty city pages with the town name swapped is not a strategy. It is a pattern Google learned to ignore a decade ago.
TL;DR Build a service-by-city matrix, but only for combinations you genuinely serve and can write about specifically. Ten real pages beat two hundred spun ones. Link them in a silo so authority flows.
The matrix, and how big it should actually be
List your services down one axis and your towns across the other. Every cell is a potential page.
| Fall River | Somerset | Swansea | |
|---|---|---|---|
| Water heater install | ✓ | ✓ | ✓ |
| Drain cleaning | ✓ | ✓ | |
| Repiping | ✓ |
Three services by three towns is nine possible pages. Resist building all nine.
Build a page only if you can answer yes to all three:
- Do we actually work in this town, regularly?
- Do we have at least one real job, photo, or customer from there?
- Can we write 300 words about this combination that would be wrong for the next town over?
If no, do not build it. An empty page dilutes your site.
The build order
Start narrow and expand on evidence.
- Your primary town plus your top service. One page, done properly.
- Your primary town plus your other services. These share local proof.
- Adjacent towns plus your top service, one at a time, as you get real jobs there.
- Everything else, only when the evidence exists.
Most service businesses top out at eight to fifteen genuinely justified pages. That is fine. That is a healthy site.
What makes a page genuinely local
The difference between a page that ranks and a template is specific detail a competitor could not copy.
- Named neighborhoods. Not “the greater area.” The Highlands, the North End, the actual names people use.
- Local housing and infrastructure reality. Housing stock age, common pipe materials, water hardness, coastal corrosion, permit process.
- Photos of your work in that town, ideally with something recognizable in frame.
- A named customer story from there.
- Local pricing context if it differs.
- Directions and landmarks. “Two minutes from the Braga Bridge.”
One paragraph of this beats a thousand words of filler. The filler is what triggers the doorway-page problem.
Page structure that works
H1: [Service] in [City], [State]
Intro: one or two sentences, service and city named naturally
H3: What we do in [City]
H3: [Local condition or problem specific to this town]
H3: Recent work in [City] ← photos, real jobs
H3: What it costs
H3: Common questions from [City] customers
CTA: phone, tappable, plus booking link
NAP + embedded map
Front-load the answer. Somebody landing here wants to know you cover them and what it costs.
Internal linking, the silo part
A silo keeps related pages linked to each other and to a parent hub, so authority concentrates instead of scattering.
The structure
- Service hub page (
/water-heater-installation/) links down to every city version. - Each city page links back up to the hub, and sideways to other services in the same city.
- Your locations page links to every city page.
- Blog posts about that service link into the hub.
Anchor text. Descriptive and varied. “Water heater installation in Somerset” is fine. Do not use the identical exact-match phrase on every link.
What to avoid. Linking every city page to every other city page. Forty pages cross-linked forty ways is a footprint, not a silo.
The duplicate content test
Before publishing, run this on any two city pages.
- Copy both into a document.
- Delete every instance of the town names.
- Read them side by side.
If you cannot tell them apart, do not publish the second one. Merge it into your service area page instead and wait until you have real material.
When a service area page is the better answer
If you cover thirty towns and have real depth in three, do not build thirty pages.
Build three strong city pages and one service area page that lists the rest in plain text. The service area page carries the coverage signal without creating twenty-seven thin pages that drag your whole site down.
Coverage and pages are different things. You can serve a town without having a page about it.
Open your two strongest city pages, delete the town names, and read them. That five-minute test tells you whether you have a content asset or a liability.
Need a pro to build these properly? [BOOK A CALL]