What a doorway page is, in Google's words
Google's spam policies describe doorway abuse as creating pages that rank for specific, similar search queries and funnel users to the same destination, and the examples include having multiple pages for different cities or regions that are essentially the same. That is the exact shape of most location page sections on small-business websites: one template, one city swapped in, repeated for every suburb in the service area.
Most businesses did not set out to build doorway pages. They were told to "build a page for every city you serve," which is good advice, and then a template was the only practical way to do it. The result is twenty pages that each say "We provide professional [service] in [City], [State]. Our experienced team serves [City] residents with..." Readers skim it, Google collapses it, and the pages compete with each other for whatever rank they do get. We see this on nearly every multi-location audit we run, which is why it is the first thing fixed in our franchise and multi-location SEO program.
The three-question test
You do not need a tool to spot the pattern. Open two of your location pages side by side and ask:
- Could I swap the city names and would both pages still be true? If yes, neither page says anything about either place.
- Is there a single sentence on this page that a competitor in that city could not also write? A named technician, a specific neighborhood, a photo from that street, a review from a customer who lives there, a note about that town's permit office or weather or housing stock.
- If a customer in that city read it, would they learn something they did not already know? Doorway pages inform nobody; they exist to be matched by a search engine.
Two "yes" answers to the first question and "no" to the other two means you have doorway pages. The checker below puts a number on it.
Try it: the location page similarity checker
Paste the body text of two location pages (select all on the page, copy, paste; the header and footer will not matter much). The checker measures how much of the wording is shared after the city names are removed, and highlights the sentences that appear on both. It runs in your browser and keeps nothing.
Paste two pages to get a score.
As a rough guide from the audits we run: under 30% shared wording is normal for two pages that genuinely describe two places (the service names and the brand will always overlap). Between 30% and 60% means the template is doing most of the talking. Above 60% is the doorway pattern, and the pages are almost certainly cannibalizing each other. The score is a diagnostic, not a Google metric; Google does not publish a threshold, and it reads structure and intent, not only words.
Why template location pages used to work and why they stopped
For years a thin city page ranked because the competition was thinner. If you were the only plumber with a "Plumber in Parker" page, you ranked for "plumber Parker" by default. Three things ended that.
Everyone built them. Once every competitor has a Parker page, Google has twenty near-identical candidates and ranks on what else it knows: the Google Business Profile closest to the searcher, the site with the most authority, the page with real content. The template stops being a differentiator and becomes a liability.
Google got better at reading. Helpful-content signals, the spam policies above and the systems behind AI Overviews all read a page for what it says, not what it repeats. A page that says nothing specific is scored as saying nothing.
AI answers want facts about the place. As we covered in our guide to Google AI Mode and local business, an AI answer is assembled from several sub-searches, and the ones about a specific town are answered by pages that know that town. A template answers none of them.
What a real location page contains
A location page that is not a doorway is a page about your business in that place. That is a different page from a page about your service with a city name on it. Here is what we put on every one we build, and the honest version of each when the business is small.
| Element | What it looks like | If you are small |
|---|---|---|
| A real headline about that place | "Water heater replacement in Parker, where half the homes are hitting the 12-year mark" | Still true, still specific; it takes one fact about the town |
| Who serves it | Named technician or team, with a photo, who actually covers that area | The owner, with a sentence about how often you are there |
| Local proof | Reviews from customers in that city, quoted with permission; photos from jobs there | One review and one photo is enough to start |
| Area specifics | Neighborhoods, HOA quirks, permit office, common housing stock, seasonal issues | Two sentences a competitor would have to research |
| Service emphasis | Which services are most requested there, with reasons | Lead with the one you do most in that town |
| Response and logistics | Travel time, scheduling windows, service fee differences if any | Honest arrival windows beat "fast service" |
| Questions specific to that place | Three to five FAQs that only make sense for that city | Even two is more than a template has |
| Structured data | Service and area markup that ties the page to the profile and the service area | Our free schema generator writes it |
Notice what is not on the list: a paragraph about how much you love serving [City], a bulleted list of every service you offer, and a map embed with nothing else. Those are the template's furniture.
The rebuild, one page at a time
You do not have to fix twenty pages at once, and you should not. Rebuild in the order of revenue at stake, and let the thin pages stay thin until their turn comes rather than deleting them; an indexed page with a little rank is worth more than a redirect.
- Rank your locations by revenue. Start where a win pays most. If you run many locations, our multi-location planner does this ordering for you.
- Collect the local raw material. Who serves the area, three jobs you did there, two reviews from residents, one fact about the town's housing or weather or permits that affects your work.
- Write the page from that material. If the raw material is thin, the page will be; that is the signal to go get more, not to pad the template.
- Add the questions only that city would ask. They come from your phone log. "Do you come out to Parker on Saturdays" is a real FAQ.
- Link it to the profile and the profile to it. If the location has its own Google Business Profile, the website field points at this page, not the homepage, and the page carries the same name, address and phone exactly.
- Re-run the checker against the next page. Two rebuilt pages should score low against each other. If they do not, the template crept back in.
A rebuilt page usually reads at 800 to 1,500 words, not because length is the goal but because real information takes room. Thin content is the symptom; the cause is that nobody collected anything to say.
When you should not have a location page at all
Some businesses have twenty location pages because someone said to, not because they serve twenty places differently. Three honest alternatives:
One strong service-area page. If you serve a metro from one address and the work is the same everywhere, one well-built service-area page with a real map, drive times and neighborhood coverage often outranks twenty thin pages and does not cannibalize anything.
A second real location. If a second city matters enough, the honest way to rank there is a presence there. Our guide to expanding to a second city covers what counts as a presence and what Google treats as a fake one.
Fewer, better pages. Five location pages that each say something true beat twenty that say nothing. The checker above is a good way to find which five.
How this fits with your Google Business Profiles
Location pages and profiles are two halves of the same signal. The profile tells Google where you are and what you do; the page tells Google, and the customer, why that matters in that place. When both exist and agree, each location ranks in its own map pack and its own organic results. When the page is a template, the profile does all the work alone, and in a competitive pack that is not enough.
If your profiles are in the same shape as your pages, start there; profile fixes move fastest, and our Google Business Profile management page has a free grader that takes two minutes per location. For the page side, the local SEO checklist lists everything a location page and its profile need to agree on. And if you want to see what a non-templated city page looks like, every one of our own service area pages was built on the rule above: a different treatment, a different module and a different set of facts for every city, because a template would have been exactly the thing this article warns about. It is the same standard we apply to our own Denver SEO presence.
Before and after: one paragraph, rewritten
The fastest way to understand the difference is to see the same paragraph both ways. Here is a typical template paragraph from a plumbing site, then the version we would write for the same company's Parker page.
Before (template): "Looking for a trusted plumber in Parker, CO? Our experienced team provides professional plumbing services to Parker residents and businesses. From water heater repair to drain cleaning, we are proud to serve the Parker community with reliable, affordable service. Contact us today for a free estimate!"
After (real): "Most of our Parker calls come from the subdivisions built between 2008 and 2014, where the original 40-gallon water heaters are now past their expected life, so we keep two on the truck when we are south of Lincoln Avenue. Mike covers Parker on Tuesdays and Thursdays and can usually be there within two hours; Saturday calls go through the on-call line. If you are in Stonegate or Pinery, you are on a septic system, and we will tell you up front if a problem is on your side of the tank."
The second paragraph is longer, but notice what it did: it named a technician, two neighborhoods, a road, a housing-stock fact, a scheduling reality and a septic detail that only applies to parts of that town. A competitor would have to know Parker to write it. A customer in Parker learns three things. And every one of the sub-questions an AI might run about a Parker plumber ("same day," "who covers my neighborhood," "do they know septic") has an answer sitting in it.
Multiply that by the number of places you genuinely serve, and you have location pages that are an asset instead of a liability. It takes an afternoon of asking your own crew questions per city. Nobody needs to invent anything.
What the problem looks like in Search Console
If you want evidence before you commit to a rebuild, Search Console will usually show the pattern in three places.
- Pages report: location pages listed under "Duplicate, Google chose different canonical than user" or "Crawled, currently not indexed." Google has read them, decided they are the same page, and kept one or none.
- Performance report, filtered by page: a handful of city pages with impressions and the rest flat at zero. The flat ones are the collapsed duplicates.
- Performance report, filtered by query: a city search where the page Google shows is the wrong city's page, or the homepage. That is cannibalization; your pages are competing and the wrong one is winning.
Any one of those is reason enough to run the checker above on the two pages involved. All three together is the full doorway signature, and in our experience the rebuild of the top three revenue locations pays for itself before the rest are done. If you would rather have the audit done for you, the free SEO audit covers location-page structure as a standard item.
Want us to do this with you?
A free audit covers your listing, your site and what AI assistants say about your category. You keep the findings either way.
Get the free audit → Call 1-800-481-8638Questions readers ask
What is a doorway page?
Google's spam policies describe doorway pages as pages created to rank for specific, similar searches that funnel users to the same destination, including multiple pages for different cities or regions that are essentially the same. A template location page with the city name swapped fits that description.
Will Google penalize my location pages?
Doorway abuse is a named spam policy, so a manual action is possible, but the more common outcome is quieter: the pages are treated as duplicates, collapse into one in the results, and compete with each other. Rebuilding them as pages that genuinely describe each place fixes both risks.
How many location pages should a business have?
As many as it can write something true and specific about. For a single-address business serving a metro, one strong service-area page often outperforms many thin city pages. For a brand with real locations, one real page per location, each linked to that location's Google Business Profile.
How do I make location pages unique without making things up?
Collect real raw material per place: who serves it, jobs you did there, reviews from residents, local housing or weather or permit facts that affect your work, and the questions customers from that city actually ask. If the material is thin, the page will be; the fix is to gather more, not to pad the template.
Should I delete my thin location pages?
Not until their replacement is ready. An indexed page with a little rank is worth more than a redirect. Rebuild in order of revenue at stake, and let the thin pages wait their turn. Delete or consolidate only the pages for places you do not meaningfully serve.