Google's doorway rule is vague about location pages. UK advertising rules are not

A page for every town, and a policy that changed its name

Open the sitemap of almost any trades business, clinic group or professional firm that has bought search advice in the past decade and you will find a layer of pages organised by place. Boiler repair in one town, then the next, then a postcode. Often they are identical but for the place name and the phone number. The owner rarely asked for forty of them.

The industry's word for the version that goes wrong is borrowed from Google: doorway pages. Google's own documentation no longer uses that phrase as a heading. The relevant section of the spam policies for Google web search is headed doorway abuse, and it opens with a definition about queries rather than places ^[1].

Doorway abuse is when sites or pages are created to rank for specific, similar search queries.

Location pages appear in the examples beneath it. One is having multiple domain names or pages targeted at specific regions or cities that funnel users to one page ^[1]. Another is creating substantially similar pages that are closer to search results than a clearly defined, browseable hierarchy ^[1]. Neither says a page about a town is forbidden. Both describe a page that exists to catch a query and hand the visitor onwards. The policy is about the relationship between a page and its destination, not about geography, and purpose is not visible in a crawl.

What the policy is actually testing

A second policy has grown around the first, and for templated location pages it is the more dangerous of the two. Scaled content abuse, in Google's wording, is when many pages are generated for the primary purpose of manipulating search rankings and not helping users ^[1]. The sentence after the definition matters more than the definition.

This abusive practice is typically focused on creating large amounts of unoriginal content that provides little to no value to users, no matter how it's created.

No matter how it is created does real work there. City pages written from a template in 2018 and pages generated by a language model last week are treated alike. The examples include using generative AI tools or other similar tools to generate many pages without adding value for users ^[1]. Volume is not the offence. Unoriginality at volume is.

Google's guidance on helpful content turns the test into questions an owner can apply with no technical knowledge. Is the content primarily made to attract visits from search engines ^[2]. Are you mainly summarizing what others have to say without adding much value ^[2]. Does it clearly demonstrate first-hand expertise and a depth of knowledge ^[2].

Applied to a page about a town those questions are easy to answer honestly. If the only thing distinguishing it from the next town's page is the place name, it was made for the search engine. If it carries the name of the person who covers that area, the response times that area gets, the parking, the local referral routes, it was made for a reader who lives there. Most location pages fail that test not because their authors were dishonest but because nobody gave them anything true to say.

What enforcement looks like when it arrives

Google states the consequence flatly: sites that violate our policies may rank lower in results or not appear in results at all ^[1].

The Search Console documentation on manual actions shows how a reviewer at Google files the complaint, and it is more useful than the policy page. There is no manual action called doorways. There is one called thin content with little or no added value, and doorway pages sit inside it, beside thin affiliate pages and scraped content ^[3]. Scaled content abuse appears as an example under a type called major spam problems ^[3]. An owner working out what happened will not find the word they were taught.

Most issues reported here will result in pages or sites being ranked lower or omitted from search results

Google adds that most reconsideration reviews can take several days or weeks, and that fixing the issue on just some pages will not earn you a partial return to search results ^[3]. For a business whose enquiries come from search, several weeks is a quarter.

The algorithmic side is unlit. Google's Search Status Dashboard records a March 2024 core update beginning on 5 March 2024 and running for 45 days, and spam updates continuing since, including ones in August 2025, March 2026, June 2026 and September 2026 ^[7]. It gives titles, dates and durations, and does not say what changed.

The British rule that is sharper than Google's

Here is the part the search industry mostly does not discuss. In the United Kingdom a second body of rules governs location pages. It is more precise than Google's and enforced by name against identifiable businesses. It is advertising law, and the Advertising Standards Authority has guidance headed ghost addresses and local claims ^[8].

It starts from a duty not to make claims, explicit or implied, that are likely to mislead consumers about the location of their offices ^[8]. It then removes the two devices most often used to manufacture a local presence. Mail forwarding addresses, which the ASA calls ghost addresses, are unlikely to be sufficient to support a claim of a presence in a particular area, and a listed address at serviced offices available for use by marketers is also likely to be insufficient ^[8].

Then it sets the test for the word itself.

would need to demonstrate a permanent registered office or branch in that area.

That is the standard the ASA says it is likely to apply to advertisers who claim or imply that they are local to a specific area ^[8]. It also closes the workaround: marketers who employ staff or have contractors in locations where they have no permanent office presence are unlikely to be able to support an unqualified local claim ^[8].

What the guidance does not do is ban location pages, and this is the sentence every multi-area business should know. Marketers wishing to communicate the availability of a service in a specific area are likely to be able to reference those locations, provided the ad does not otherwise imply, by inclusion or omission, the existence of physical local premises ^[8]. Serving a town is fine. Reading as though you are in it is not.

The impression can be created with no explicit claim at all: some marketing techniques can create the impression that the marketer is a small local business ^[8]. That is a finding about design and tone, assessed on the overall impression the advertising is likely to give consumers ^[9]. The rule behind it is one line. Marketing communications must not materially mislead or be likely to do so ^[9].

Five rulings, and the line they draw

The guidance is abstract. The rulings are not. In October 2019 the ASA upheld a complaint about www.wandsworthplumbers.com, run by Property Improvement & Maintenance Services Ltd. The home page was headed plumbers for Wandsworth, Southfields, Earlsfield, SW18 and referred to the local professionals at Wandsworth Plumbers ^[10]. The complainant understood the business to be based in east London. The ASA found consumers would read that as an indication of a permanent base in Wandsworth, and that it had seen no evidence of one ^[10].

In July 2024 it went after a network. Four locksmith websites, trading as Gauge B-Ready Locks, Guardian Tech Locksmith Cranleigh, ApexCraft Locksmith Frimley and Alex the Locksmith West Molesey, each showed a single contact address, two with photographs of a shopfront carrying the business name ^[11]. The ASA said it had not seen any evidence that a locksmith business operated from any of the addresses featured in the ads, and understood them to be residential homes or unrelated businesses ^[11]. The sites shared the same design and text and said they had been built by the same person ^[11].

Two rulings show the line is tighter than most businesses assume. In April 2019 Leftmove Estate Agents Ltd was found to have misled by listing a Poulton branch among twelve locations when what it had was a serviced office with a receptionist. The ASA acknowledged the services were provided by that office, then found that they did not have a fully operational branch or property agents based in the area ^[12]. Real work, real address, still misleading, because the word used was branch.

In December 2021 Ben's Gutters Ltd was found to have misled through presentation rather than through any claim the ASA found false. The leaflet was styled as a compliments slip, with apparently handwritten text about cleaning a neighbour's gutters, a request to please call me, and a hand-drawn smiley face ^[14]. The company was a larger operation working in various parts of the country. The ASA found the overall impression was of a small, independent, local business, and was therefore misleading ^[14].

The counterweight is Purplebricks, instructive because the ASA did not uphold it. In October 2017 it considered the claim find your Local Property Expert and concluded that the word local would be generally understood to refer to an individual's expertise and knowledge of the area, rather than their physical location ^[13].

Together the five say something specific. Claiming to cover an area is defensible and provable. Claiming, or designing a page to suggest, that the business sits in that area is neither, unless it does. The failing was never the page about a place. It was the implied premises.

The statute behind the self-regulator

It would be easy to treat that as self-regulation with no teeth. The reading is out of date. The unfair commercial practices provisions of the Digital Markets, Competition and Consumers Act 2024 are the backdrop now, and the Competition and Markets Authority's guidance states that they apply to practices taking place from 6 April 2025 ^[15].

Section 226 covers the provision of false or misleading information, and section 227 the omission, or the unclear, untimely or obscure provision, of material information ^[15]. The guidance describes the average consumer as reasonably well informed, reasonably observant and reasonably circumspect, and notes that there need not be an actual transaction for there to be a transactional decision ^[15]. That last point is the relevant one for a page that generates an enquiry rather than a sale.

Two cautions belong here. The ASA's advice pages state expressly that they are CAP Executive advice, not legal advice, and do not bind CAP or the ASA ^[9]. And this article states no penalty figures under the 2024 Act, because the guidance on penalty levels was not verified for this piece.

Why the Business Profile is the harder boundary

Most of the commercial pressure behind location pages is not about pages. It is about the map. A page can be written about anywhere. A Business Profile cannot.

Eligibility begins with physical reality: a business that either has a physical location customers can visit, or travels to customers where they are, can create a Business Profile ^[4]. The staffing condition is explicit, and it catches most schemes.

your business location should be staffed by your team and able to receive customers during its stated hours.

From there the guidelines close the doors one by one. PO boxes or mailboxes located at remote locations aren't acceptable ^[4]. A physical postal address the business does not operate out of, also known as a virtual office, is not eligible ^[4]. A co-working desk qualifies only if the office has clear signage, receives customers during opening hours and is staffed by the business's own staff ^[4]. Service-area businesses should hold one profile for the central office with a designated service area, and hide the address from customers ^[4]. Breaking the guidelines can lead to suspension, or removal of business information from Google ^[4].

So the thing a business most wants from a location page, which is to appear in the map results for a town it is not in, is the one thing a location page cannot deliver. Google says local results are mainly based on relevance, distance and popularity, defines distance as how far each business is from the customer who is searching, and warns that there is no way to request or pay for a better local ranking on Google ^[5]. Distance is measured from a real address.

What a location page can honestly carry

None of this makes the location page a bad idea. A business with three sites needs three pages, because the sites have different addresses, hours, staff and services. A single-site business with a travel radius needs something else: a page that explains the service and states the area honestly.

The test that survives both Google's policies and the ASA's rulings is whether the page contains information only someone who works there could have written. In practice that is a short list of things that differ by location and are true.

• The address and hours of the site itself, matching the Business Profile, since Google asks for business hours, including special opening hours, to be complete and accurate ^[5].
• The named people who cover that location and what they are qualified to do, the first-hand expertise Google's content guidance asks about ^[2].
• What differs there: services offered at that site, waiting times, access, parking, transport, the relationships specific to the area.
• The area served, stated as area served, with no photograph, address block or map pin implying premises the business does not occupy ^[8].
• Structured data describing what is visible on the page, using the LocalBusiness vocabulary, which supports hours and departments within a business ^[6].

The negative list is shorter. Do not publish a page for a town the business does not serve. Do not give each town page its own telephone number to look like a branch network. Do not put an address or shopfront photograph on a page for a place where there is no shop, the pattern the ASA found misleading in the locksmith ruling ^[11]. Do not call a serviced office a branch ^[12].

If a page cannot be given anything true and specific to say, the right decision is not to write it better. It is not to have it.

What it means for a local business

For an owner with an existing set of location pages, the work is an audit rather than a rewrite.

1. List every page that targets a place. Open each beside another and mark the sentences that differ other than the place name. If nothing does, that is the pattern Google calls substantially similar pages and unoriginal content at scale ^[1].
2. Ask whether a reader would finish each page believing the business has premises there. If so and it does not, that is the ASA's implied premises problem, the version of this risk with named precedents attached ^[8][10].
3. Check the pages against the profile. A business may hold one profile per real location and cannot hold one at a virtual office or remote mailbox ^[4]. A page supporting a listing that should not exist is a liability before it is a page.
4. Merge or delete the pages with nothing of their own, keep those tied to a real site, and give the remainder one honest page stating the area served. Record what changed and when, so later movement in enquiries can be read against it.

The commercial case is not the avoidance of a penalty, which may never come. It is that the exercise reveals how little of the content was ever capable of persuading anyone. Pages that say nothing specific do not convert either.

For regulated businesses the calculation is sharper. A clinic, practice or firm advertising a location it does not occupy is exposed twice, once to the advertising rules and once to whatever its professional regulator says about how services may be described. The advertising risk is the one that arrives by letter with a published ruling attached, and ASA rulings are public and durable in search results.

What remains uncertain

The settled parts of this subject are the documented ones. Google's policy wording, its manual action categories, its Business Profile eligibility rules, the ASA's guidance and the published rulings all say what they say, and anyone can read them.

What nobody outside Google can establish is how, or how often, the doorway abuse and scaled content abuse policies are applied to ordinary small-business location pages. Google publishes no breakdown of manual actions by type and no threshold at which similar pages become substantially similar. The dashboard logs dates and durations and nothing else ^[7], and Google says it keeps search algorithm details confidential to make the ranking system as fair as possible for everyone ^[5]. Any consultant who says how many town pages is too many is guessing.

Nor is there a published controlled study, as far as this article's research found, isolating the effect of adding or removing location pages on enquiries for UK local businesses. The evidence is documentary, not experimental.

The regulatory side has open edges too. The 2024 Act's regime has applied only since 6 April 2025 ^[15], a short period in which to judge how the CMA will treat location claims by small traders. One gap is worth naming: a 2026 ASA ruling on local claims by an estate agency network was identified during research but could not be retrieved, so it is not relied on here.

What is not uncertain is the direction of the test. Platform and regulator ask the same question in different words, which is whether the page corresponds to something real. Google asks whether the content was made for a reader or for a query ^[2]. The ASA asks whether the overall impression matches the facts the business can evidence ^[9]. A location page that would satisfy an inspector standing in the doorway satisfies both. One that exists only in a sitemap satisfies neither, and the reason to remove it is not fear of enforcement. It is that it was never doing the work it was sold as doing.