If you sell a service across a county, the advice you'll hear sooner or later is to build a page for every town you cover. Web design in Ashford, web design in Canterbury, web design in Folkestone, and so on down the map. More pages, more chances to rank. It sounds like simple maths.
I tried it on my own site. At one point Business Sorted Kent had around 700 pages built from the same handful of templates: every service crossed with every town I could reasonably drive to. It looked thorough. Google didn't agree.
What actually happened
Search Console told the story plainly. The number of pages Google chose to index fell from 385 in early August to 257 by early October, and 226 by late October. Hundreds sat in a report called "Crawled, currently not indexed" (311 in the most recent export). That wording matters. Google had found the pages, read them, and decided they weren't worth showing anyone.
Nothing was broken. There was no penalty, no technical fault, no hack. Google simply looked at a page about web design in Herne Bay, compared it with my page about web design in Whitstable, saw the same paragraphs with a different town name, and concluded that one of them was plenty. Often it decided none of them were.
That's the part most location page advice leaves out. Google doesn't reward you for the number of pages you publish. It rewards pages that give someone something they couldn't get elsewhere, including elsewhere on your own site.
Why near-identical pages hurt more than they help
There are three problems with templated town pages, and they compound.
They compete with each other. If ten pages say essentially the same thing, Google has to pick one to show, and it won't always pick the one you'd choose. Your strongest page ends up sharing its signals with nine weaker copies.
They spend Google's attention. Every site gets a finite amount of crawling. Hundreds of thin pages mean Google spends its visits on pages that will never rank, instead of the ones that might.
They lower the impression your whole site gives. Google has said for years that low-value content in bulk can weigh on how it sees a site overall. A site that's 80 per cent filler looks like a site that's mostly filler, because it is.
What I kept, and why
When I rebuilt the site, I didn't keep town pages on gut feel. I went to the Search Console data and set a rule: a town page stayed only if the old version had earned more than 1,000 impressions, or had a real search query sitting in the top 20 positions with at least 10 impressions behind it. Evidence that Google already associated me with that town.
Ten towns passed: Ashford, Canterbury, Folkestone, Gillingham, Herne Bay, Maidstone, Margate, Ramsgate, Sittingbourne and Tunbridge Wells. Everything else was redirected to the nearest page that genuinely answered the same question, so nobody following an old link hits a dead end.
Then I rewrote those ten from scratch, one at a time.
What a location page needs to earn its place
A useful test: if you swapped the town name for another town, would the page still be true? If yes, it isn't a location page. It's a service page with a place name pasted in.
Each of my remaining town pages now has things that only apply to that town:
- How far it is from me and how I'd get there. I'm based in Ashford, so Canterbury and Maidstone are a short drive and Margate is a longer one. That's real, and it changes how a visit works.
- What people there actually search for. Pulled from my own Search Console data, not guessed.
- The local business picture. Business counts for the district from the ONS, with the source and year cited, so the page says something true about the market.
- Real local proof where it exists. Client work in or near that town. Where there isn't any, the page doesn't pretend.
- Questions people from that town ask. Answered properly, not padded.
Each page ended up somewhere between 700 and 1,000 words, and I checked them against each other so no two pages share more than a small fraction of their wording.
It's slower than generating 700 pages. It's also the only version that works.
How to tell if your site has this problem
If you have Search Console set up, open the Pages report and look at "Crawled, currently not indexed". If that list is full of your town or service pages, Google is telling you it doesn't think they add anything.
Then read three of your location pages side by side. Out loud, if you can bear it. If they sound the same, they are the same, as far as Google is concerned.
The fix isn't to delete everything. It's to decide which places genuinely matter to your business, make those pages properly useful, and point the rest somewhere sensible with redirects.
The uncomfortable bit
A lot of web design and SEO companies still sell location pages by the hundred, because they're cheap to make and the number looks impressive in a proposal. I understand the appeal. I fell for it myself, on my own site, and I do this for a living.
Fewer, better pages is a harder sell. It's also what Google keeps rewarding.
If you'd like a second opinion on your own location pages, I'm happy to take a look. You can see how I approach SEO for small businesses, or get in touch and tell me what you're working with.
Questions people ask about location pages
Should I have a page for every town I cover? Only for towns where you have something specific to say and some evidence of demand. A short list of strong pages beats a long list of weak ones.
Will deleting location pages lose me traffic? If a page gets real traffic, keep it and improve it. If it gets none, removing it and redirecting it to a relevant page usually costs nothing and helps the pages that remain.
How long should a location page be? Long enough to be genuinely useful for that place. In practice, the pages that do well tend to be several hundred words of specific detail, not 150 words with the town name swapped in.