Every few months a service business decides it needs to rank "near me" and does the obvious thing: it multiplies its page count. One page for every service, times every city in the state, times every suburb, until a five-person studio in Mohali has 240 pages claiming "interior designers in Zirakpur," "interior designers in Kharar," "interior designers in Derabassi."

It never works the way the spreadsheet promised. Google's spam policies name this pattern directly: doorway pages — multiple near-identical pages aimed at specific queries that funnel visitors to the same destination — are a violation, not a tactic (Google Search Central, spam policies). And even short of enforcement, a page with no unique information has nothing to rank for except the city name in the title.

The fix is not "write better city pages." The fix is to stop treating service areas as pages at all, and treat them as structured data that a smaller number of honest pages can render.

1. Decide the coverage model before writing any template

Two different business types get conflated constantly:

  • A fixed-location business. Customers come to you. Your address is your coverage. One LocalBusiness entity, one physical address, hours, a map.
  • A service-area business. You go to customers. The address may be a home office you don't want published. Your coverage is a set of named areas or a radius, and that set is a business fact, not an SEO choice.

Write the answer down in a file the whole team can read. If the owner can't name the areas where they will actually send a crew next month, the correct area-page count is zero.

A useful test before adding any area to the list: could you describe, in one sentence, a job you completed or could take tomorrow in that area? If not, it is not coverage. It is a keyword.

2. Put the coverage in JSON-LD, where it can be validated

LocalBusiness has separate fields for where you are and where you serve. Use them for what they mean:

<script type="application/ld+json"> { "@context": "https://schema.org", "@type": "ProfessionalService", "name": "Example Studio", "telephone": "+91 00000 00000", "url": "https://www.example.com/", "address": { "@type": "PostalAddress", "streetAddress": "4th Floor, F135 Phase 8B, Industrial Area, Sector 74", "addressLocality": "Sahibzada Ajit Singh Nagar", "addressRegion": "Punjab", "postalCode": "160055", "addressCountry": "IN" }, "areaServed": [ { "@type": "City", "name": "Mohali" }, { "@type": "City", "name": "Chandigarh" }, { "@type": "City", "name": "Panchkula" } ] } </script> 

Two rules that catch most mistakes:

  • address is a fact, areaServed is a claim. Never list an area in areaServed that has no operational answer behind it.
  • Do not emit a separate LocalBusiness node per city. One entity, many areas. Multiple location nodes with the same phone and different cities is the machine-readable version of a fake branch office.

Validate whatever you ship with the Schema Markup Validator and check that the rich-results test reads the same values you intended.

3. Make uniqueness a build-time requirement, not a review note

Area pages can exist — but only where the page has something the service page doesn't. In practice that means the template should refuse to render unless a minimum set of fields is populated:

Field Why it must be unique
Local project or job reference Proves work happened there
Travel or scheduling note Proves the area is operational, not decorative
Local constraint (permit, society rule, terrain, typical property) Proves local knowledge
Named contact responsible for that area Proves ownership

If you publish through a generator, put the gate in the generator. A cheap pre-deploy check looks like this:

// Reject an area page that is essentially the service page with a city string swapped. const cities = ['Mohali', 'Chandigarh', 'Panchkula', 'Zirakpur', 'Kharar']; const strip = (s) => cities .map((c) => new RegExp(`\\b${c}\\b`, 'gi')) .reduce((acc, re) => acc.replace(re, ' '), s) .replace(/<[^>]+>/g, ' ') .replace(/\s+/g, ' ') .trim(); const base = strip(await fetch(serviceUrl).then((r) => r.text())); const page = strip(await fetch(areaUrl).then((r) => r.text())); const overlap = page.length / Math.max(base.length, 1); if (overlap < 0.55) { throw new Error(`Area page is ${Math.round(overlap * 100)}% of the service page with the city removed`); } 

The threshold is arbitrary on purpose. The point is that the check exists and fails the build, so "we'll add proof later" cannot ship.

4. Keep the URL and canonical graph boring

  • Area pages live under a stable prefix, e.g. /areas/mohali/, not /best-interior-designer-in-mohali-cheap/.
  • Every page is self-canonical. Canonicalising every area page to the service page tells Google the area pages are duplicates — because, structurally, they are.
  • Titles follow one pattern: Service in Area | Brand. No chain of four cities and three adjectives in one title.
  • The XML sitemap lists what you want crawled and nothing else. If an area page is too thin to include, it is too thin to exist.
  • Internal links come from the service page's coverage section and the footer of the area page's neighbours — not from 240 footer links across the site.

5. Do not outsource the honesty to the review system

A service-area page that promises same-day response in an area you reach in three hours is a content problem that becomes a reviews problem. Google prohibits discouraging or prohibiting negative reviews, and review gating — filtering customers before asking for feedback — is explicitly out of bounds. The legitimate version is dull: ask every customer, publish everything, respond to everything.

6. A pre-deploy checklist that takes ten minutes

  1. List every city token in the built site: grep -rio "mohali\|chandigarh\|panchkula" dist | wc -l. Count should match the areas you can actually serve.
  2. For each area page, compute the overlap check above and record the number.
  3. Validate the JSON-LD and confirm areaServed equals the coverage list, not the keyword list.
  4. Fetch each area URL and confirm 200, correct canonical, and a title in the agreed pattern.
  5. Check the sitemap count against the number of pages you would defend in a meeting.
  6. Search the site for the office phone and address and confirm they match the Google Business Profile exactly.

That is the whole audit. It is not exciting, and that is the point: "near me" rankings are decided by proximity, prominence and relevance signals that a page factory cannot manufacture. What a page factory can do is create a thin, duplicated surface that wastes crawl budget and gives a reviewer a reason to look at you.

If the honest coverage list is three areas instead of thirty, publish three pages and make them worth reading. That is a smaller site that can actually be maintained.

Design Den Solutions is a design and web development studio in Mohali, Punjab. We build the pages and the structured data described here — see web development. This is our own referral link, not third-party coverage.

Written with AI assistance and reviewed before publishing. Contact: +91 98721 90965 · www.designdensolutions.com