What Is Structured Data?
Schema markup is machine-readable code, usually JSON-LD, that tells Google and AI search engines exactly who your business is: name, address, phone, hours, license, service area, and services. It is not a direct ranking factor, but it qualifies your pages for rich results and makes your business dramatically easier for AI systems to verify and recommend.
Your visible page says "Call us for bathroom remodeling in Portland." Your schema says, in a format with no room for interpretation: this is a HomeAndConstructionBusiness, named exactly this, at this address, with this phone number, this license, these hours, serving these cities.
Humans infer. Machines parse. When a crawler has to infer your service area from a paragraph of marketing copy, it gets it wrong or gives up. When your schema states it outright, there is nothing to infer. That is the entire point of structured data: it removes ambiguity about the facts of your business.
What Schema Does a Local Business Need?
The foundation is LocalBusiness, and the first decision is the subtype. Schema.org defines specific types for the trades: GeneralContractor, RoofingContractor, Plumber, Electrician, HousePainter, and the broader HomeAndConstructionBusiness they sit under, plus equivalents for nearly every other local category. Choosing the most specific subtype that describes you tells search engines what you are, not just that you exist.
Google's own documentation requires surprisingly little for LocalBusiness, name and address, but the recommended list runs to fifteen-plus properties, and the recommended properties are where the value lives: telephone, openingHoursSpecification, geo coordinates, areaServed, sameAs links to your profiles, priceRange, and your real NAP stated character-for-character the same way it appears everywhere else on the web. On top of that foundation, a complete local graph adds Service schema for each thing you sell, BreadcrumbList so Google understands your site structure, WebSite and Organization to tie the entity together, and FAQPage where you genuinely answer common questions.
These nodes should reference each other, forming a connected graph rather than five unrelated snippets. A Service node that points back to the LocalBusiness that provides it, in the cities that business serves, is how a machine builds a confident picture of your company. This connected-graph approach is the core of our technical SEO and schema engineering work.
What Schema Does, and What It Does Not Do
Here is the honest version that most schema guides skip. Structured data is not a ranking hack. Google has said for years that schema is not a direct ranking factor, and adding it to a weak site will not lift you past stronger competitors. What it does is qualify you for enhanced treatment: rich results like breadcrumbs in your search listing, eligibility for review stars where Google's rules allow them, and a knowledge panel that treats your business as a verified entity rather than a string of text.
Eligibility is not entitlement, and the rules move. Google restricted FAQ rich results to well-known government and health sites back in 2023, so your FAQPage schema is unlikely to render as an accordion in search results today, even though it still helps machines parse your answers. Review markup has its own strict rules about self-serving reviews. And even perfectly valid markup only makes you eligible; Google decides what to show. Anyone who promises specific rich results from schema alone is selling you certainty Google does not offer.
So why bother? Because understanding compounds. Schema makes every other signal you send, your reviews, your citations, your content, your Google Business Profile, easier for machines to attribute to the same entity. It is the difference between Google assembling a picture of your business from scraps and you handing it the labeled photograph.
Does Schema Matter for AI Search?
The newer reason to take structured data seriously is the one most agencies still underweight: AI search engines read it too. When ChatGPT, Perplexity, or Gemini tries to answer "who does bathtub refinishing near Beaverton," it leans on the same structured facts Google does: entity identity, services, location, service area. A business with a complete schema graph is dramatically easier for an AI system to describe accurately, and an AI system that understands you confidently is far more likely to mention you.
This is not speculative futurism; it is how retrieval-augmented answers work today. The AI systems cite and summarize sources they can parse cleanly. Thin or broken schema makes your business harder to parse, which makes you easier to skip. In 2026, schema markup for a local business is as much about being legible to AI as it is about rich results in classic search.
Why Page Builders Ship Broken or Thin Schema
If your site runs on Wix, Squarespace, Weebly, or a themed WordPress install, your schema was generated by the platform, and platform-generated schema is a recurring character in our audit findings. The typical problems: a generic LocalBusiness type with no trade-specific subtype, missing areaServed and geo data, stale information that no longer matches the Google Business Profile, and template junk that marks up things the page does not actually show, which violates Google's guidelines outright.
A real example: when we audited Rip City Construction's Squarespace site, the auto-generated schema listed the wrong ZIP code. The site had been silently disagreeing with the company's own Google Business Profile about where the business was, for years. Nobody had ever looked, because page-builder schema is out of sight and out of mind. That is the pattern: the platform checks a box labeled SEO, the owner assumes the box means something, and the actual markup goes unexamined until someone runs a validator on it.
How Do You Test Your Own Markup?
You can check your site in about ten minutes with two free tools. Google's Rich Results Test shows whether your pages qualify for enhanced search features and flags errors in the markup it finds. The Schema.org validator (validator.schema.org) checks your full JSON-LD against the vocabulary itself, catching type mistakes and malformed properties the Google tool ignores. Run both against your homepage and your most important service page.
Then do the human check the tools cannot: open the JSON-LD (view source, search for ld+json) and read what it claims. Does the business name match your signage and your Google Business Profile exactly? Is the address current? Does the phone number match the one on your contact page? Is the type specific to your trade, or a generic Organization? If you serve twenty cities, does areaServed say so? A schema block that validates perfectly while stating outdated facts is worse than none, because it is confidently wrong.
In Google Search Console, the enhancement reports show which rich-result types Google has detected across your site and whether any have errors worth fixing. It is worth a quarterly look, because markup that was valid at launch drifts as content changes.
Where to Start
If the checks above surfaced thin, generic, or contradictory markup, you have two options. Patch it by hand, which is viable on a small site if you are comfortable editing JSON-LD, or treat it as part of the technical foundation the next time the site gets rebuilt properly. Either way, the standard to hold is a connected graph: specific LocalBusiness subtype, real NAP, geo, license numbers where you have them, areaServed, sameAs, plus Service and Breadcrumb nodes that reference it.
If you would rather have someone else read the markup, our free ECARAT visibility audit includes a structured-data assessment: we parse your schema graph, check it against your Google Business Profile and citations, and benchmark the result against five local competitors. You get a scored report of exactly what your site tells machines today, and what to fix first.