How to Create Location Pages That Actually Rank

Most location pages fail for a single reason: they are one template with the town name swapped. Google names that pattern doorway abuse in its own spam policies, and it treats the mass-produced version as scaled content abuse. Both have been demotion triggers for years, not fresh risks that arrived with the last update.

This guide is the build manual, not the pep talk. You get an ordered checklist you can run yourself, a plain answer on how many location pages you actually need (usually a fraction of what an agency will quote you), a one-line test for whether a sentence belongs on the page, and two popular tactics worth dropping. If you are still deciding whether you can outrank a national chain at all, start with our companion piece on how to beat big brands in local search. This page assumes you have decided, and want pages that survive.

Reading time: about 13 minutes.

Why templated city pages fail

Templated city pages fail because Google's spam policies target them by name. The doorway-abuse policy describes sites or pages created to rank for specific, similar search queries that lead users to intermediate pages less useful than the final destination, and it gives one example that is the exact tactic: having multiple pages targeted at specific regions or cities that funnel users to one page. A second policy, scaled content abuse, covers many pages generated for the primary purpose of manipulating rankings and not helping users. Between the two, the entire city-swap playbook is described in Google's own words. There is no clever version of it that the policy did not anticipate.

The damage is not confined to the thin pages either. When a large share of a site is near-duplicate, the pattern lowers trust in the whole domain, so the pages you actually care about inherit the drag. And the classifier does not read intention. A page you built to be helpful still counts as duplication if the only thing separating it from the next town is a find-and-replace on the place name. This is why a business can add forty location pages and watch its existing rankings soften rather than climb.

How many location pages you actually need

You should have one location page for each place you genuinely operate, and almost never more. Genuinely operate means a real address you staff, or a defined area you actually work in with people, projects, and knowledge on the ground. A plumber with one shop and a service radius has one, perhaps two honest location pages, not thirty. The number an agency proposes is often set by what is billable, not by how many places you can write about truthfully, and that gap is where the doorway risk is manufactured.

BrightLocal's guidance is a useful floor: aim for most of each page to be unique to that location, with 40 to 60 percent unique content the minimum that justifies a separate URL. If you cannot clear that bar for a town, that town does not get a page. It gets a paragraph on a page you can defend. The same firm frames the decision as a question worth writing on a sticky note: would this page still be useful if search engines did not exist? If the honest answer is no, you are building for a crawler, which is precisely the intent the doorway policy exists to catch. The wider distinction between chasing many geographies and owning a few is something we lay out in local SEO versus national SEO, because getting the scope wrong is the most expensive early mistake we see.

Here is the count decision worked all the way through. Say you are a contractor who takes jobs across eight towns in one county. Two of them, the town you are based in and the neighbour where half your work happens, you can describe in real detail: named subdivisions, the permit office you actually deal with, jobs on specific streets. Those earn a page each. The other six, where you have done the occasional job but hold no real local knowledge, do not. They get named honestly inside your main service-area copy and your coverage map, with no dedicated URL pretending to an expertise you do not have. Two defensible pages beat eight thin ones, and the six honest mentions still catch the searches that matter.

The build: a page that cannot be copy-pasted

A location page that survives is built from elements a competitor structurally cannot reproduce. Work down this list for the one page you are building first, and do not open a second until this one is genuinely done. Each row is something you can do yourself, with no agency required.

Notice what is not on the list: a paragraph of generic service description, a stock photo of a handshake, and a block of keywords. Those are the parts a template ships by default, and they are the parts that earn nothing. The whole value of a location page lives in rows two through six, the material only you have.

If one element feels abstract, start with the job stories in row three, because they are the easiest to get right and they carry the most weight. The shape is plain: what the customer called about, where in the area it was, what you found, and what you did about it. Four sentences, a real street or neighbourhood, and a photo of the actual work. Three of those on a page tell a search engine and a reader more than any volume of service description, and none of them can be lifted onto a page for the next town.

The copy-paste test

Here is a one-line test that catches a doorway page before Google does. Take any sentence on your draft, swap the town name for a neighbouring town, and ask whether the sentence is still true. If it is, that sentence is not local content. It is boilerplate wearing a place name, and it is doing none of the work you built the page for.

Run two real examples through it. "We provide reliable roofing services to homeowners in Paris, Ontario" survives the swap to any other town unchanged, so it earns nothing. "The 1960s bungalows along the older streets of Paris were built with roof decks that fail a specific way in freeze-thaw, and we stock the fasteners that job needs" breaks the instant you change the town, which is the entire reason it works. The second sentence is something a national competitor's legal review would never sign off, and something a resident reads as true. That is the target for every line on the page.

If you already run a stack of location pages, apply the test to what you have before you write anything new. Open three of them side by side and read the equivalent paragraph on each. If those paragraphs differ only by the place name, you have found your doorway risk, and the fix is not to reword them into synonyms, which the scaled content policy also covers. The fix is to either fill each page with real local material or collapse the weakest into one honest page and redirect the rest. Fewer strong URLs almost always outrank a wider set of thin ones.

Two tactics to stop doing

Two widely repeated location-page tactics do not work, and the evidence against them is not subtle. Both survive because they are easy to sell and easy to bill, not because they move rankings.

Correcting these two matters more than repeating the usual advice, because both cost you the scarce resource that actually builds the page: the hours you would otherwise spend gathering real local material. Time spent geotagging a photo library is time not spent writing down what happens on a job in that town.

What makes the page citable, not just rankable

The same specifics that beat the doorway filter are what get your page quoted by AI answers. Generative engines assembling a local answer need something concrete and attributable, and the Princeton-led team that coined Generative Engine Optimization found content-level changes could raise a page's visibility in generative responses by up to 40 percent, with cited statistics, quotations, and authoritative references among the methods that performed. Read that against a national brand's location page: real prices, named conditions, and dated jobs are exactly the elements a corporate legal review removes, and exactly what an independent operator can add this week.

So the non-templated page is not a separate AI project bolted onto the SEO work. It is the same page. The section on the failure mode of a specific soil type, the paragraph naming what a repair really costs in your city this year, the review that names a street: those are the passages a model can lift and attribute. Corporate copy written to be safe across four hundred markets contains almost nothing a machine can quote, which is the opening an independent business should be walking through.

The honest effort and timeline

Building one location page properly takes a few focused hours, not a weekend of spinning text. Most of the time goes into gathering the real material: pulling two or three job stories, getting a quote from a local customer, confirming the pricing and conditions that genuinely differ in that place. The writing itself is fast once the facts are in front of you, because you are recording what you already know rather than inventing filler. Ranking then follows the ordinary local curve, which is measured in months and depends on your trade, how competitive the town already is, and how many honest pages you can sustain over time. The tactical groundwork that shortens that curve, from Business Profile to schema, is covered in how to rank for near me searches.

For a sense of what disciplined local content can do from a standing start, consider a documented cold start rather than a promise. A brand-new foundation repair company in a major Alberta metro, on a fresh domain with no reviews and no search history, went from zero to 30 leads inside its first three months (internal lead tracking, 2026). The engine was entity groundwork and locally specific pages built on that metro's real soil and freeze-thaw conditions, not a bank of templated town pages. We publish that case anonymised, with the client name shown on the audit call rather than on the open web where a competitor would read it. The qualifier belongs in the same breath as the number: lead volume swings hard by trade, market, and season, and foundation repair is high-ticket enough that thirty enquiries is a genuine pipeline where it would not be for a café. Our Results Guarantee is written for existing domains, since a brand-new domain's timeline is honestly longer, and pretending otherwise would be the kind of claim this page argues against.