Most Indian service businesses I audit have schema markup on the site. Very few have schema that does anything. There is a difference between having a JSON-LD block in the head and having Google treat that block as a description of a real entity it can verify.
Here is what separates the two in 2026.
The rich results that still exist for service businesses
Google has quietly retired or de-emphasised a lot of rich result types. For a service business in India, only a handful reliably render now:
- FAQ rich results — still showing, but heavily restricted to authoritative domains. If your site has no established trust, marking up FAQs gets you the validation tick and nothing in the SERP.
- Review snippets — only when the review is about a product or a service the page itself is about, and only when the review data is visible on the page. Self-serving review markup on an "About Us" page is a manual action waiting to happen.
- Breadcrumbs — the least glamorous and the most consistently rendered. Almost every service site should have these and most don't.
- Local business knowledge panel enrichment — not a SERP feature exactly, but LocalBusiness markup that matches your Google Business Profile strengthens the entity association.
Everything else — HowTo, most Event variants, the old sitelinks searchbox — has either gone or narrowed to the point where it isn't worth engineering time.
The mistake: markup that contradicts the page
The single most common failure I see is schema that says something the page does not say. A LocalBusiness block listing an address that appears nowhere in the visible HTML. An aggregateRating of 4.9 with no reviews rendered. An openingHours specification on a site with no hours anywhere.
Google's structured data guidelines are explicit that markup must describe content visible to the user. When it doesn't, one of two things happens: the markup is ignored, or the site gets flagged. Neither is the outcome you paid for.
The fix is mechanical. Open the rendered page, open the JSON-LD, and check every property against something a human can see. If a property has no visible counterpart, delete it or add the counterpart.
Entity consistency is the part nobody does
Schema is how you tell a search engine what kind of thing you are. But a single JSON-LD block on a single page is a claim, not evidence. Evidence is the same claim repeated consistently across the web — your site, your Google Business Profile, your directory listings, your author bios, your social profiles.
If your site says "Deep Bhardwaj, SEO consultant, New Delhi" and your directory listings say "DB Digital, Delhi NCR", you have given the engine two weak entities instead of one strong one. Pick one name, one service description, one city, and keep them byte-identical everywhere. This matters more every year as AI-generated answers lean on entity graphs rather than page-level relevance.
The sameAs property is the mechanism. On your Person or Organization block, list the canonical URLs of every profile that describes the same entity. That is how you connect the dots for a crawler that otherwise has to guess.
Practical implementation order
For a typical Indian service site with 20 to 60 pages, this is the sequence that produces results:
- Organization or Person block on the homepage, with
name,url,logo,address,telephone, and a completesameAsarray. One canonical block, one page. - BreadcrumbList on every page below the homepage. Cheap, universally rendered, and it helps Google understand your site hierarchy.
- Service blocks on each service page, with
serviceType,areaServed, andproviderpointing back to the homepage Organization by@id. The@idreference is what turns a set of isolated blocks into a graph. - FAQPage only where real questions are answered in visible copy. Three to six questions. Not fifteen keyword-stuffed pseudo-questions.
- LocalBusiness only if you have a physical address customers can visit. If you're remote-first, use Organization and skip the address claims.
Validate with the Rich Results Test, not just the Schema Markup Validator — the latter tells you the syntax is legal, the former tells you whether Google can actually do anything with it.
What to expect
Schema does not make a page rank. It makes an already-relevant page easier to interpret and more likely to be shown as something richer than a blue link. For a business competing on service keywords, the realistic gain is better click-through on positions 3 to 8 and a stronger entity signal feeding into AI overviews.
That's worth doing properly. It isn't worth doing as a checkbox.
If you want this done as part of a broader technical audit, this is the kind of work an SEO expert in India should be handling as baseline, not as an upsell.
Related: I wrote earlier about how AI search engines choose which Indian websites to cite, which covers the entity side of this in more depth.