City Page Duplicate Content: How Similar Is Too Similar?
TL;DR: There’s no duplicate content penalty, which is the fact most people get wrong in both directions. What actually happens to near-identical city pages is worse in a quieter way: Google picks one, ignores the rest, and you can’t tell why your forty pages produced four rankings. The practical threshold is around 40% unique body content, and there’s a way to test it before you commit to the whole set.
Table of Contents
- What Google actually does with near-duplicates
- Why there’s no penalty and it still hurts
- A workable similarity threshold
- What has to change per city
- Reusing sections legitimately
- Testing before you publish forty pages
What Google Actually Does With Near-Duplicates
Google’s documentation is consistent on this: duplicate content on a site is generally not grounds for manual action unless it’s deceptive. What happens instead is a selection process.
When Google finds multiple URLs with substantially the same content, it chooses one as canonical and consolidates signals to it. The others may be crawled but not indexed, or indexed and effectively never served. In Search Console you’ll see them under “Duplicate without user-selected canonical,” “Alternate page with proper canonical tag,” or most commonly “Crawled – currently not indexed.”
Three things to understand about that behavior:
Google chooses, not you. It may not pick the city page you care most about. It often picks whichever accumulated a few links or got crawled first.
The unselected pages produce nothing. Not reduced value. Effectively zero.
No error is reported. Nothing appears in any report labelled as a problem. Your site looks healthy and forty of your pages are inert.
That’s why this survives on so many sites. There’s no alarm.
Want more customers from Google & AI search?
Get a free SEO audit of your site — see exactly what to fix first.
Why There’s No Penalty and It Still Hurts
Four real costs, none of which look like punishment.
Wasted production budget. You paid for forty pages. If eight got indexed, you paid five times the effective rate.
Signal splitting. If several pages do get indexed and each earns a fragment of relevance for the same query, none accumulates enough to rank well. This is the same problem as keyword cannibalization, arriving from a different direction.
Crawl effort spent on nothing. Real but usually minor at local-site scale.
Diluted topical clarity. A site that’s 70% near-identical city pages is harder for any system — search engine or answer engine — to characterize. What is this site authoritative about? Hard to say. It mostly repeats itself.
The framing that matters for a business owner: you didn’t get penalized, you got ignored. Ignored is more expensive because it’s invisible.
A Workable Similarity Threshold
No official number exists. Anyone quoting one as Google policy is inventing it. What follows is a practical operating threshold that holds up in the field:
| Unique body content | Typical outcome |
|---|---|
| Under 20% | Almost certainly not indexed. Google treats these as one page. |
| 20 – 40% | Coin flip. Some indexed, some not, mostly unstable. |
| 40 – 60% | Generally indexed. Ranks if there’s real demand. |
| Over 60% | Behaves like independent pages. |
Aim for 40% minimum, target 50%+.
How to measure it: take the body text of two city pages — excluding header, nav, and footer, which are boilerplate on every page and shouldn’t count in either direction. Compare the two texts. Any diff tool, plagiarism checker, or text-similarity tool gives you a percentage. Compare each new page against its two most similar siblings, not against a random one.
One nuance worth knowing: it’s not purely word count. A page that shares 70% of its wording but leads with two paragraphs of genuinely specific local information can behave better than the percentage suggests, because the distinctive material is prominent and above the fold. Position matters, not just proportion. Put the unique content first.
What Has to Change Per City
Ranked by how much difference each element actually makes.
High impact — do these:
Job-specific narrative. Two or three paragraphs about work you’ve actually done in that city. Conditions, timing, what the job required. This is the strongest differentiator available and the one competitors can’t copy.
Photos from that city. Original job photos. Different per page. If you have geotagged photos from actual jobs, better still.
Reviews from customers in that city. Even two, with first name and neighborhood.
Local conditions that change the work. Housing age, water hardness, soil, terrain, typical construction era. For a plumber in a 1920s neighborhood versus a 2015 subdivision, the actual work differs, and describing that difference proves you know the area.
Named sub-locations. Neighborhoods, subdivisions, main roads, landmarks. Not a keyword-stuffed list — used naturally in sentences.
Medium impact:
Response time and dispatch detail from your nearest location, with the honest drive time.
City-specific FAQ. Permit requirements, municipal water authority, local code quirks. Genuinely differs between jurisdictions and few competitors bother.
Local competitive positioning. What’s underserved in that market.
Low impact — necessary but not sufficient:
Title, H1, and meta description with the city name. Do it, but understand it changes almost nothing about the duplication problem. Swapping the city name in the title of otherwise identical content is precisely the pattern that gets ignored.
Reusing Sections Legitimately
Some repetition is correct. Nobody expects your certifications to be different in each city.
Fine to repeat verbatim:
– Licensing and insurance details
– Your service process, if it genuinely doesn’t vary
– General safety or preparation information
– Contact mechanics, hours, payment methods
– Company history
Should vary:
– Everything in the high-impact list above
– Any claim about response time or coverage
– Service emphasis, where local demand differs — flood response leads in a low-lying area, frozen pipes lead at elevation
A structure that works:
- Opening: 200–300 words, entirely unique. Local conditions, a real job reference, why this city has particular needs. Unique content first, always.
- Service detail: mostly templated, with one or two city-specific sentences worked in.
- Local evidence: entirely unique. Photos, reviews, neighborhoods served.
- Process and credentials: templated. Fine.
- City FAQ: mostly unique. Permits, local authority, jurisdiction quirks.
- CTA: templated with the local phone or dispatch detail.
That produces roughly 45–55% unique content with a manageable amount of writing per page — and critically, the unique material occupies the opening and the middle rather than being buried at the bottom.
Testing Before You Publish Forty Pages
The whole point. Do this and you’ll never waste a forty-page build again.
Step 1 — Build three, not forty. Pick your three highest-priority cities: most job history, most demand.
Step 2 — Measure similarity between them before publishing. Under 40%, rewrite. This is cheap now and expensive later.
Step 3 — Publish all three, internally linked from a relevant service or service-area page. Add to sitemap.
Step 4 — Wait six weeks. Genuinely wait. Not two.
Step 5 — Check three things in Search Console:
– Are all three indexed? (Pages report, or URL Inspection per page.)
– Is each earning impressions for its city query? (Performance, filtered by page.)
– Do any of them show impressions for another city’s query? That means Google is confusing them, which is a signal to increase differentiation.
Step 6 — Read the result honestly:
| Result | What it means | Next move |
|---|---|---|
| All indexed, all earning | Template and depth work | Build the next batch of 5–10 |
| Indexed, no impressions | No demand, or too weak to compete | Reassess the city list before building more |
| Some indexed, some not | Right at the similarity threshold | Increase uniqueness on the unindexed ones, retest |
| None indexed | Too similar, or site authority too low | Stop. Fix differentiation. Do not build forty. |
Six weeks and three pages to learn whether your approach works, versus six months and forty pages to learn it didn’t. That’s the entire argument.
Quick Recap
- There’s no duplicate content penalty. Google selects one page and ignores the others, silently.
- Being ignored costs more than being penalized because nothing reports it.
- Target 40% minimum unique body content, 50%+ preferred, measured excluding nav and footer.
- Job narratives, city photos, local reviews, and local conditions are the differentiators that matter. City names in titles are not.
- Put unique content first — prominence matters, not just proportion.
- Build three pages, wait six weeks, read the indexing result, then decide whether to build forty.
Frequently Asked Questions
Will Google penalize my city pages?
No. It will select one and ignore the rest, which produces the same practical outcome as a penalty without any notification.
Can I use the same photos across city pages?
You can, and it weakens every page. Original photos per city are among the cheapest differentiators available to a service business — your crews are already taking them.
How do I check whether my pages are considered duplicates?
Search Console → Pages, and look at “Crawled – currently not indexed” and the duplicate-related statuses. URL Inspection on an individual page will also tell you whether Google selected a different canonical.
Should I noindex weak city pages?
Better to improve or consolidate them. If a page can’t be made distinct and there’s real demand, fold it into a regional page covering several nearby towns.
Is a regional page better than several city pages?
Frequently yes, especially for adjacent small towns. One strong page covering a county or metro area with genuine local depth usually outperforms eight thin town pages, and it’s far less work to maintain.
Does this apply to service pages too?
Yes, and it’s just as common. Five service pages describing near-identical processes with a swapped service name have exactly the same problem.
Keep Reading
Want more customers from Google & AI search?
Get a free SEO audit of your site — see exactly what to fix first.
