Schema markup is the single piece of technical SEO work that most small business websites get wrong, mostly because nobody bothered to explain it in plain language. This is that explanation.
What schema actually is
Schema is structured data — small blocks of standardized code that sit inside your webpage and tell search engines and AI systems exactly what the page is about. Instead of forcing a machine to interpret your About page from prose alone ("hmm, this seems to be about a wellness studio in Florida"), schema explicitly says "this is a LocalBusiness, type Yoga Studio, located in St. Petersburg, with these hours, these services, these reviews."
That removal of ambiguity is the whole point. Machines that are not guessing about your page are far more likely to surface it correctly.
Schema turns "we think this page is about X" into "this page is about X."
Why it matters more every quarter
Schema has been around for over a decade. For most of that time, it was helpful but not essential — Google could usually figure out a well-written page from prose alone. Two things changed that recently.
First, rich results. Schema is what powers the review stars, FAQ accordions, recipe cards, and other visual features that make some search results stand out from the rest. These directly raise click-through rates from the search engine results page.
Second, AI answer engines. When ChatGPT, Google AI Overviews, Gemini, or Perplexity synthesize an answer, they preferentially cite pages with structured data because the data is unambiguous. Two equally good pages on the same topic — one with schema, one without — the one with schema gets cited more often. This effect is growing as AI answer experiences take more of the search market.
The schema types that matter for small business
Schema.org defines hundreds of types. Most small businesses need fewer than seven. Here's what to focus on.
Essential
LocalBusiness (or a more specific subtype)
The foundation. Identifies your business, location, contact info, hours, and service area. Use a more specific subtype when one fits — Restaurant, Dentist, LegalService, HealthAndBeautyBusiness, ProfessionalService. More specific subtypes give the machine more context.
Where: homepage, in the head as JSON-LD.
Essential
FAQPage
For any page that contains a series of question-and-answer pairs. This is the highest-leverage schema for AI citation in 2026 — answer engines pull FAQ content directly into their synthesized answers. Add 5–8 real questions with substantive answers on your homepage and major service pages.
Where: any page with FAQ-style content, including the homepage.
Essential
BreadcrumbList
Tells search engines and AI systems how a page fits into your site's hierarchy (Home → Field Manual → This Article). Powers the breadcrumb display in search results and helps machines understand site structure.
Where: every page that sits inside a section (not the homepage).
Content
Article
For blog posts, guides, and editorial content. Includes author, publication date, modification date, and main image. Especially important for AI engines, which weight content with clear authorship and recency signals.
Where: every blog post, guide, or editorial page.
Content
HowTo
For step-by-step instructional content. Each step gets its own structured block with name and text. AI engines disproportionately cite HowTo schema when answering "how do I..." queries, which are a large share of all questions asked to ChatGPT and Gemini.
Where: any page that walks through a sequence (a checklist, a tutorial, a setup guide).
Service-based
Service
For individual service pages. Identifies what the service is, who provides it, where it's offered, and what it costs. Many small businesses skip dedicated service pages entirely; for those who don't, Service schema makes those pages markedly more competitive.
Where: any dedicated service or offering page.
E-commerce
Product and Offer
For e-commerce sites with discrete products. Product schema is what enables the price, availability, and review-star display in search results — and what makes a product likely to be cited in shopping-focused AI answers. Skip if you sell services rather than products.
Where: every product page on an e-commerce site.
Authority
Organization or Person
Used to identify the entity behind the website itself — the business as an organization, or a notable individual associated with it. Connects your site to a unique entity that search engines and AI systems can triangulate across the rest of the web (social profiles, directories, press mentions).
Where: typically embedded inside the LocalBusiness schema as a reference, or as its own block on About pages.
Where the markup goes
JSON-LD schema markup lives inside a <script type="application/ld+json"> block. The script can go in the document head (typical for organization-level schema) or inside the body (for page-specific schema). Multiple schema blocks per page are allowed and often appropriate — a single article page might carry Article schema, BreadcrumbList, FAQPage, and HowTo simultaneously.
Common mistakes
Three patterns to avoid. First, plugin-generated schema that doesn't match the page. Many WordPress and Shopify plugins auto-generate schema with stale or generic data. Always review what's actually being output. Second, schema that says something different from what the page says. If the schema lists hours as 9–5 but the page says 8–6, Google flags the inconsistency. Schema must match visible content. Third, over-stuffing — adding schema for things that aren't actually on the page (fake reviews, services you don't offer, prices you don't charge). This is treated as deception and can result in penalties.
How to check it works
Google's free Rich Results Test at search.google.com/test/rich-results validates schema and shows which rich result features the page qualifies for. The Schema.org Schema Markup Validator at validator.schema.org does broader validation. Both should run clean before deployment. Anything that throws errors in those tools won't earn you any of the visibility benefits schema is supposed to provide.