Get Townsmith

Topical authority for local businesses: the service-and-area model

Topical authority for a local business is earned on two axes at once: how completely you cover your service, and how genuinely you serve each area. Volume of pages is not the signal. Coverage backed by real, distinct content is. A pile of thin near-duplicate location pages does not build authority – it spreads your trade thin across empty pages and trips Google’s spam policies, so the honest route is to deepen a few service topics and prove genuine local presence in each place you actually work.

This is the strategy hub that the rest of our local-SEO guides hang from. It ties together how you link pages together and how many location pages you should actually have, so read it first and follow the links down into the specifics. Written by Stephen Evans, maintainer of Townsmith. Last reviewed 30 June 2026.

What is topical authority for a local business?

Topical authority is the trust a site earns by covering a subject completely and credibly, so that search engines and the people reading treat it as a reliable source on that subject. For a national publisher that is one axis: subject depth. For a local business it is two axes held at the same time. The first is service depth – you genuinely cover your trade, from the common jobs to the awkward edge cases. The second is area depth – you genuinely serve each place you claim, with real evidence of having worked there.

Most local sites are strong on one axis and hollow on the other. A boiler-repair firm might write one excellent service page and then bolt forty town pages onto it that say the same thing with the place name swapped. That is deep on service and empty on area. The reverse also happens: forty genuine local case studies with no underlying explanation of what the work involves. Authority comes from filling both axes with content that could not have been produced by find-and-replace.

Google does not have a separate setting for local authority. It rewards the same thing everywhere: helpful, people-first content. In its “Creating helpful, reliable, people-first content” guidance (Google Search Central, current as of 2026), the self-assessment asks “Does the content provide original information, reporting, research, or analysis?” and “Does your content clearly demonstrate first-hand expertise and a depth of knowledge (for example, expertise that comes from having actually used a product or service, or visiting a place)?” For a local trade, visiting the place is the literal job. That first-hand experience is the raw material competitors copying each other cannot fake.

Does page count build authority?

No. Page count is not a ranking signal and never has been. Coverage is the signal, and a hundred near-identical pages cover nothing – they cover one thing a hundred times. Worse, scaling thin pages is a named risk in Google’s own spam policies, so the page count that was meant to build authority can be the thing that removes it.

Two policies bite here. Google’s spam policies (Google Search Central, last updated 15 May 2026) define scaled content abuse as “when many pages are generated for the primary purpose of manipulating search rankings and not helping users,” and name “using generative AI tools or other similar tools to generate many pages without adding value for users” as an example. The same document defines doorway abuse as “when sites or pages are created to rank for specific, similar search queries” and lists “having multiple domain names or pages targeted at specific regions or cities that funnel users to one page.” A set of location pages that exist only to catch “[service] in [town]” searches and shuttle everyone to one contact form is the textbook example.

The trap is that thin location pages feel like progress. The page count goes up, the sitemap grows, and for a while it looks like coverage. But each page dilutes the next, the cluster reads as machine-made, and the whole portfolio drags. For the full picture on how many pages a real service-area business should publish and why, see our guide on how many location pages you actually need. The doorway failure mode is set out in detail in what doorway pages are and how to avoid them.

Authority-building versus doorway behaviour

The line between a real local page and a doorway page is not about the topic. Both target “[service] in [town].” The difference is whether the page earns its place. Here is the same intent done two ways.

  • Builds authority: one strong service hub explaining the trade in depth, with area pages that add genuinely local detail – jobs you have done there, parking and access quirks, the housing stock, local response times.
  • Doorway behaviour: dozens of area pages spun from one template, identical but for the town name, every one funnelling to the same enquiry form with nothing local on the page.
  • Builds authority: publishing only the areas you actually cover, and saying plainly which towns you do and do not serve.
  • Doorway behaviour: publishing a page for every town within fifty miles whether you work there or not, to catch the search.
  • Builds authority: a small set of supporting guides that explain the work and link down to the relevant service hubs.
  • Doorway behaviour: keyword-stuffed pages that read as a list of search phrases rather than something written for a person to read.

If you removed the town name from one of your area pages and it could be any town, it is doorway behaviour. If removing the town name leaves a page that no longer makes sense because the local detail is woven through it, you are building authority.

How do you build the two axes at once?

You build service depth and area depth as one structure, not two projects. Pick a small number of core service topics, give each a proper hub, support them with a handful of guides, and connect everything with descriptive internal links so the cluster reads as one body of knowledge.

Pick the core service topics first

Start from what you actually do, not from a keyword list. A heating engineer’s core topics might be boiler repair, boiler installation, and annual servicing. Each gets one hub page that covers the topic in real depth – the symptoms, the typical fixes, what a fair job looks like, what it costs to do properly. These hubs are the spine. They are where service depth lives, and they are what your area pages and guides point at. The distinction between a service hub and a location page is worth getting right before you build either: see service pages versus location pages.

Add area depth only where it is real

Area pages carry the second axis, and they only work when there is something genuinely local to say. A page for boiler repair in Uckfield should reference the older terraced stock off the high street, the response time from your base, the jobs you have actually completed there. A page for Lewes should read differently because Lewes is different. If you cannot write a paragraph about a town that nobody could have written without working there, you are not ready to publish that page yet. Hold it back rather than padding it. For the running list of what genuinely local content looks like, see multi-location SEO.

Prove experience, expertise, authoritativeness and trust

Both axes feed the same underlying quality Google looks for: experience, expertise, authoritativeness and trust. For a local trade, experience is the photographs of your own work, the named engineer, the registration number, the real address. None of this is decoration – it is the proof that you are a business that has been to these places and done this work. We cover how to demonstrate it on a local site in E-E-A-T for local businesses.

How do clusters link up and down?

A topical cluster is held together by links that flow in two directions. Supporting guides link down to the service hubs they explain. Area pages link up to this strategy hub and across to the relevant service hub. The hub pages link out to the areas and guides that support them. Done well, a reader or a crawler can move between any two related pages in a click, and the shape of the linking tells search engines which pages are central and which are supporting.

  • Down: a guide on signs your boiler is failing links down to the boiler-repair service hub.
  • Up: the Uckfield area page links up to this topical-authority hub and to the boiler-repair hub it serves.
  • Across: the boiler-repair hub links across to boiler installation and servicing, because a real customer often needs the next thing.
  • Descriptive anchors: link with the words a person would search for, never “click here” or “read more.”

The internal-linking mechanics – which pages link to which, what the anchors should say, how to avoid orphan pages – are involved enough to have their own guide. Read internal linking for local SEO before you wire a cluster together, because a strong structure linked badly still underperforms.

A worked example: a heating engineer covering the Sussex Weald

Picture a one-van heating firm based near Uckfield, covering a ring of towns across the Sussex Weald. The wrong instinct is to publish a page for every town the van could theoretically reach – thirty pages, all spun from one template. The right structure is smaller and earns more.

  • Three service hubs: boiler repair, boiler installation, annual servicing. Each is a few hundred words of genuine explanation, written once, properly.
  • Four area pages, not thirty: Uckfield, Lewes, Crowborough, Heathfield – the towns where the firm has actually completed jobs and can write something true about each. Each links up to this hub and across to the service hubs.
  • Three supporting guides: “what a boiler service should include,” “no hot water: what to check first,” “is your boiler worth repairing or replacing.” Each links down to the service hub it supports.

Ten genuine pages beat thirty hollow ones because every page in the smaller set carries weight. When the firm later starts working in Hailsham for real, it writes the Hailsham page then – with real local detail – rather than having published an empty placeholder months earlier. The discipline of publishing only what is real is the whole game, and it is the opposite of the scaled-content failure mode.

Is there a special trick for AI search?

No. The two-axis model is the optimisation. What gets a local business cited by an AI engine is the same thing that ranks it in classic search: a clear, self-contained answer near the top of each page, real depth on the service, and genuine proof you have done the work locally. There is no AI-only shortcut that rescues a thin page, so the discipline is identical – cover the topic properly and earn the local detail.

Schema helps a machine understand a page it can already read – it does not rescue a page with nothing to say. On a service-area business the useful pattern is LocalBusiness with Service and areaServed, where the service area in the markup mirrors the towns named in the visible text exactly. Never invent a street address or phone number for a town you only visit. Get the markup right and it confirms what the page already says.

Where Townsmith fits

Townsmith is the mechanism for building this structure honestly. It builds real, editable location pages with proper internal links and clean schema, all generated on your own server with no AI and no external calls in the free plugin. Its quality score rates each page from 0 to 100 against the depth and distinctiveness this guide describes, and it holds thin pages back rather than letting you publish a near-duplicate that would read as doorway behaviour. The free Safety Check gives an advisory verdict on how doorway-prone your portfolio looks – guidance, not a guarantee from Google – and Twin Finder flags pages that are too alike.

It complements your existing SEO plugin rather than replacing it. If you run Rank Math, Yoast, AIOSEO or SEOPress, those keep owning your meta and base schema, and Townsmith builds the pages. If you want to see how the engine and the quality score work, the features page walks through it. The point is not to publish faster. It is to publish only the pages that are real, which is exactly what this two-axis model asks of you.

Common questions

How many service topics should a local business cover?

As many as you genuinely offer, and no more. A firm doing three things well should have three strong hubs. Inventing topics to look broader dilutes the site the same way thin area pages do. Depth on a few real services beats shallow coverage of services you rarely deliver.

Will fewer, better pages really outrank a competitor with hundreds of location pages?

Often, yes, and increasingly so. Competitors leaning on scaled near-duplicate pages are exposed to the doorway and scaled-content policies in Google’s spam policies, and core updates have repeatedly hit that pattern. Page count was never the ranking signal. A smaller set of pages that each cover something real is the more durable position.

Can I use AI to write my location pages?

Google judges the result, not the tool. Its spam policies name generative AI used “to generate many pages without adding value for users” as scaled content abuse, so AI that mass-produces near-duplicate pages is the risk. AI that helps you organise facts you already know, while you supply the real local detail and judgement, is a different thing. The free Townsmith plugin uses no AI at all and makes no external calls. Its optional Pro assistant only works from facts you type, never invents specifics, and never publishes on its own.

How do I know if a page is too thin to publish?

Apply the town-name test: remove the place name and see whether the page still says anything specific. If it reads as a template, it is too thin. Hold it back until you have done a real job there and have something true to write. Townsmith’s quality score and Twin Finder are built to flag exactly this before a page goes live.

Where to go next