Search "local SEO suburb pages" and you will find a hundred articles telling you to build a page for every suburb you service. I read those articles. Then I built forty of them for my own website.
Four months later the site had no organic rankings worth the name and almost no search traffic. Not slow progress. Nothing.
So I audited my own site the same way I audit a client's, and what I found was uncomfortable enough that I deleted thirty-four pages in an afternoon. This is what happened, with the real numbers.
The theory, and why it is so convincing
The logic seems airtight. Somebody in Coburg searches "electrician Coburg". Google likes pages that match the search. So if you have a page that says "Electrician Coburg" you should win that search. Do that for every suburb you cover and you own the whole map.
It is convincing because it is almost right. Google genuinely does reward location relevance, and businesses that never mention their suburbs genuinely do lose local searches. The part everyone skips is the bit about what those pages have to actually contain.
What I actually built
Forty suburb pages out of sixty-one pages on the site. Fitzroy, Richmond, Brunswick, Coburg, Preston, Box Hill, Doncaster, Dandenong, Werribee — the lot. Each around four hundred words. Each with the suburb name in the title, the heading, the copy and the URL.
They were not spun garbage. I wrote them. They read fine. If you landed on the Richmond page you would not think "this is spam" — you would think "this is a page about Richmond".
That turned out not to matter, because nobody reads them one at a time. Google reads them all at once.
The number that ended the argument
When I ran a similarity comparison across the forty pages, the average overlap was 87.6%. But the number that actually landed was this one:
Words that were identical between my Fitzroy page and my Richmond page. Two suburbs on opposite sides of the river, and nearly two-thirds of the text was word-for-word the same.
The only things that changed were the suburb name, two street names, and one rewritten sentence. I had written a template and filled in a blank forty times without ever admitting to myself that was what I was doing.
Here is the test I now use. Put two of your suburb pages side by side and delete the suburb name from both. If you cannot tell which is which, Google cannot either — and it will pick one and ignore the rest.
What Google calls this
There is a name for it in Google's own spam policies: doorway pages. Google defines them as sites or pages created to rank for specific searches that then funnel users to the same destination. Multiple pages targeting close variations of a query, where the user ends up somewhere generic, is the textbook example.
I was not trying to spam anyone. That is worth saying, because most people who build these pages are not either. But intent is not what gets measured. Similarity is.
The part that surprised me most
I assumed the worst case was that the suburb pages simply would not rank. That would have been annoying but survivable — forty pages doing nothing, the rest of the site fine.
That is not what happens. Two-thirds of my site was near-duplicate content, and that became the site's overall signal. The blog posts and service pages I had actually put work into — eleven hundred words each, genuinely researched — were sitting in a neighbourhood of thin repeated pages. They did not rank either.
Search Console showed it plainly once I knew where to look. Under Pages, two categories mattered:
- "Duplicate, Google chose different canonical than user" — Google had looked at my suburb pages, decided they were the same page, and picked one to represent the group. The rest were discarded.
- "Crawled — currently not indexed" — Google visited, read the page, and decided it was not worth adding to the index at all.
Those two categories accounted for eighteen pages between them. Not blocked, not broken, not penalised in any dramatic way. Just quietly judged and set aside.
What I did about it
I deleted thirty-four suburb pages and kept six.
The six I kept are the suburbs I can actually write about honestly — places I have worked, visited, or can drive to this afternoon and photograph. Every deleted page was 301-redirected to the areas overview page first, before anything was removed, so nobody lands on a 404 and any signal those URLs had carried transfers rather than evaporating.
Order matters more than people realise. Redirects first, deletion second. Do it the other way round and you create thirty-four live 404s while you work, which is exactly the wrong signal to send while Google is already unsure about your site.
Then I resubmitted the sitemap — down from sixty-one URLs to twenty-eight — and waited.
What I would build instead
I have not abandoned suburb pages. I have abandoned forty suburb pages. There is a version of this that works, and it looks like this:
- Three to six suburbs, not forty. Pick the ones where you actually want work and where you can say something true.
- A thousand words minimum, and genuinely different ones. If the page could be about anywhere, it is about nowhere.
- Real local detail. Street names you have worked on. The housing stock — Californian bungalows in one suburb, sixties walk-ups in another. Parking. Which jobs come up most in that area and why.
- Photos from that suburb. Your own, not stock. This is the single hardest thing to fake and the most convincing when it is real.
- A different reason to exist. The Coburg page should answer a question the Preston page does not.
The test is whether somebody who lives there would recognise the place. If a Brunswick local reads your Brunswick page and thinks "yeah, that is here" — you have built something. If they could not tell it from your Preston page, you have built a liability.
If you already have thirty of these
Do not panic-delete them this afternoon. Work in this order:
- Compare two of them properly. Paste both into a diff tool. Look at the actual overlap rather than trusting your memory of writing them.
- Check Search Console → Pages. If "Duplicate, Google chose different canonical" or "Crawled — currently not indexed" have numbers against them, Google has already told you what it thinks.
- Decide which handful you can genuinely rewrite — the suburbs that actually matter to your business.
- Redirect the rest to a single areas page. 301, not 404, and set the redirects up before you delete a thing.
- Then rewrite the survivors properly. This is the real work, and it is the part everyone skips.
The honest summary
Four months of ranking for nothing was not caused by anything technically broken. There was no noindex tag, no robots.txt problem, no slow server — page speed was measured at well under a quarter of a second to first byte. The site was fine.
It was caused by me doing what the articles said, at volume, without asking whether the pages were worth reading. Forty mediocre pages did more damage than four good ones would have done good.
If you take one thing from this: the number of suburb pages you have is not the metric. The number a local would recognise is.