Get Townsmith

Location pages for multi-location dental practices, without the duplicate-content penalty

Each practice page must differ in substance, not just the town name: named clinicians, the treatments actually offered at that site, accepted plans, access and parking, and real local detail. Pages that only swap the place name get filtered as duplicates.

Dental groups feel this sooner than most. A practice in Reading and a practice in Basingstoke share the same brand, the same treatment list and, very often, the same page template. Build forty of those and the search engine treats most of them as one. This page is not about the duplicate-content argument itself, which is settled and linked below. It is about the dental-specific substance that makes a practice page genuinely its own, and the check the Townsmith engine runs before it will let you publish.

What has to genuinely differ on each practice page

The brand, the booking system and the broad pitch can be shared. The following details are practice-specific, cannot be transferred from a sibling page, and are what separate a real location page from a name-swap stub:

  • The named clinicians who actually work at that site, with their roles and GDC registration numbers. The General Dental Council number is public, verifiable and unique to a person, which is exactly the kind of detail a content mill will not bother to attach to the right practice.
  • The treatments genuinely offered at that practice. Not every site does implants, clear aligners, sedation or facial aesthetics. List what this one does, and do not list what it does not.
  • The plans and cover accepted there: NHS availability and whether the list is open, the private membership or payment plan in use, and which insurers are processed. NHS versus private is the single biggest practical question a patient has, and it varies by site.
  • Access and getting there: on-site or nearby parking, step-free access and ground-floor surgeries, the nearest station or bus route, and opening hours including any late or weekend clinic.
  • Real local detail from that catchment: how the practice handles registrations and emergencies for that town, languages spoken by the team, and the kind of patient it actually sees, written from experience rather than a template.

None of this is treatment advice, and that is deliberate. This is page structure and honesty: a page that carries true, checkable, per-site facts reads as its own page to both a patient and a crawler.

The swap-the-town-name test

Here is a thirty-second check you can run on any practice page before it goes live. Read the page, then mentally swap the town name and the practice name for another branch’s. If the page would still read as true and complete after the swap, it is too thin: nothing on it belongs to that specific practice. A page that fails the swap, because it names the wrong clinicians, the wrong plans or the wrong parking, is a page that is actually doing its job.

Why the duplicate risk is real, and where to read the argument

Near-identical practice pages get folded together: the search engine keeps one, drops the rest, and you lose the catchments you built the pages for. The full mechanics, how sibling similarity is measured and what thresholds bite, are set out in our guide to duplicate content across location pages, and the wider estate-of-pages strategy sits in the multi-location SEO guide. We will not re-argue it here.

The policy footing is Google’s, not ours. Mass-producing pages that add little value for users is named in Google’s spam policies as scaled content abuse (Google Search Central, spam policies, accessed 4 August 2026). The scaled content abuse policy was introduced in the March 2024 spam update. Pages that exist only to capture a town name, with no substance behind them, are also the textbook definition of a doorway page. For dental groups the operative E-E-A-T member is Trust, earned by honest, verifiable per-site content, not by any health claim.

What a real practice page proves that a stub cannot

A genuine practice page is hard to fake because the proof is non-transferable. Naming Dr A Patel (GDC 234567) as the principal dentist at the Reading practice, alongside a hygienist and a treatment co-ordinator who actually work there, ties the page to people the GDC register can confirm. Stating that this site holds an open NHS list for under-18s but takes adults privately on a monthly plan, while the Basingstoke site is private-only with a separate insurer, is a fact a patient acts on and a fact that cannot be copied to a sibling without becoming false.

This is also why a page’s honesty decays. A page that was genuinely distinct in March can drift towards its siblings by August as a clinician leaves, an NHS list closes, or a treatment is added at one site and not another. Substance is not a one-off task; it is the thing that keeps the page distinct, and the thing worth re-checking. The E-E-A-T for local business guide covers how this experience and trust read to a search engine, and the local landing page checklist covers the per-page mechanics.

How the quality score stops you publishing a name-swap stub

This is the part you can see for yourself. When you generate a practice page in Townsmith, the engine scores it out of 100 before you publish, and it does it without any AI and without making a single external call. Everything runs in PHP on your own server. The largest share of the score, 45 points, comes from sibling similarity: every page’s visible text is broken into five-word shingles and compared against its closest sibling using Jaccard overlap. A further 35 points come from local substance, the filled slots and area-specific detail described above, and 20 from baseline quality such as original text, a real H1 and a meta description. The full scoring breakdown is in the quality score documentation.

So a second practice page that is 90% identical to the first, a name-swap stub, scores badly on similarity and substance and lands below the default publishing threshold of 60. In the free plugin the engine warns you: it shows the finding, for example “88% similar to the Reading practice”, as an advisory gate so you cannot publish a duplicate by accident without seeing it. In Pro the engine can hold the page back as a draft until you raise it. The Twin Finder and Safety Check tools surface the same problem across an existing estate. The number is something you act on, not a feeling.

Pro also includes an optional AI assistant for filling the per-site facts, and it is off by default. When you turn it on it works only from facts you type in, the clinicians, plans and access for that practice. It never invents a clinician, a plan or a patient outcome, and it does not write a page for you. It cannot promise a ranking, and no tool can promise compliance: the substance has to be true, and only you can confirm that.

Schema for the practice pages

This guide carries Article schema referencing the Rank Math base graph plus a BreadcrumbList, matched to the visible text on the page. The practice pages themselves are a different shape: each one should emit LocalBusiness with its own name, address and phone, plus Service entries for the treatments it actually offers, all mirroring what is visible on that page. The pattern, and the merged worked example, are in the local business schema guide, Service plus areaServed section.

One thing to leave off: do not add Review or AggregateRating markup to your own practice pages to try to win star ratings. Google has not shown self-serving review stars since 2019, so the markup earns nothing and risks a structured-data manual action if it does not match visible content. Stars in results come from product markup and third-party platforms such as Google, not from rating yourself.

Keeping the pages honest over time

Publish the pages a tranche at a time rather than all at once, and watch how each one performs before leaning on the next. Our guide to measuring location pages in Search Console covers the index-coverage and performance checks worth running, and how many location pages to build helps you decide which catchments deserve a page at all. If you would rather the scoring, similarity check and substance prompts ran for you on every save, that is what the Townsmith engine does: see the features.

Common questions

Can multiple dental practices share one page template?

Yes. A shared template is normal and sensible, and it is not what causes a duplicate-content problem. What causes the problem is shared substance: when the visible text barely changes from one practice to the next. Keep the structure shared and make the clinicians, plans, treatments and access genuinely different on each page.

Should I publish a page for a practice that only differs by NHS availability?

NHS versus private is a real difference and a genuine reason patients choose one site over another, so it is worth stating clearly. On its own it is rarely enough. Pair it with the named team, the actual treatment list, access and parking for that site, and the page becomes its own. A page whose only distinct line is the NHS status will still read as a near-duplicate.

Is listing clinicians and treatments a YMYL medical-claim risk?

Listing who works at a practice, their GDC registration, and which treatments are offered is factual page structure, not treatment advice. It stays out of medical-claim territory because it does not promise outcomes or recommend a treatment for a condition. Trust is built by being verifiable and honest, which is the safe side of the line for a dental site.

Will the quality score guarantee my pages rank?

No. The score is a pre-publish check that the page is substantive and distinct from its siblings; it stops you shipping a thin name-swap. It is not a ranking promise, and no plugin can make one. What it does is remove the most common self-inflicted reason multi-practice pages fail, which is that they were near-identical to begin with.

Written by Stephen Evans. Published 4 August 2026. Scoring figures describe the Townsmith engine’s default configuration and are current as of 4 August 2026.