What Types of Schema Does My Business Need?

Schema markup is one of the most important AI visibility levers. It is also one of the most confusing because the schema.org vocabulary lists hundreds of types and the official documentation does not tell you which ones actually matter for your business. After running hundreds of FlinnSchema audits, the pattern is clear: most businesses need somewhere between five and ten schema types, and the right mix depends almost entirely on what kind of business you run. This guide walks through which schemas every business should have, then breaks down the additions by business type with real-world examples from client work.

If you would rather see the full technical reference first, our complete schema markup guide for 2026 covers every type with code examples. This post focuses on the decision-tree question of what your specific business needs.

Why Schema Choice Matters More Than You Think

The temptation, especially after reading a long schema guide, is to implement every schema type that might possibly apply. This is a mistake. Each schema type has its own validation rules and required fields. Implementing schema poorly is often worse than not implementing it at all, because invalid schema can cause AI engines and Google to ignore your entire JSON-LD block.

The correct approach is to identify the schema types your business genuinely needs, implement them completely with all required and recommended fields, and validate them. Five well-implemented schema types beat fifteen partial ones every time. Our audits consistently show that the businesses with the highest AI visibility scores are not the ones with the most schema types, but the ones with the most complete schema across the types that actually fit their business.

Schema also varies in how much AI engines care about each type. Some types like Organisation and LocalBusiness are heavily weighted because they describe the business entity itself. Others like ItemList or HowTo are useful in specific contexts but skipped entirely by AI engines for most queries. Choosing the wrong types wastes implementation time and crowds out the work that would actually move your score.

The Universal Schemas Every Business Needs

Regardless of what kind of business you run, three schema types form the foundation. These are the schemas that AI engines and Google use to understand who you are at the entity level. Without these, your other schema types lose context.

• Organisation is the most important schema for every business. It identifies your brand as a specific entity, includes your name, logo, URL, social profiles, contact information, founding date, and address. Place this in the head of your homepage and reference it from other schema blocks via @id. AI engines use Organisation schema heavily when building their understanding of who you are.
• WebSite describes your overall site, including the SearchAction property that tells AI engines how your site search works. This is a small but high-value addition because it makes your site searchable from inside AI tools that support the property.
• BreadcrumbList on every non-homepage URL tells AI engines and Google how your site is organised. It is structural rather than informational but it is one of the easiest schemas to add and it has consistent value.

These three together give AI engines a clear picture of your brand, your site structure, and your search behaviour. If you do nothing else, do these three. They are the foundation everything else builds on. The official reference for each type lives on Schema.org, and Google publishes its own structured data requirements in the Google Search Central documentation.

E-Commerce Businesses (Shopify, WooCommerce, BigCommerce, Custom)

If you sell products online, the schemas that matter beyond the universal three are:

• Product on every product page, with name, description, SKU, brand, image, price, availability, and ideally GTIN or MPN. Missing Product schema is the single biggest gap we find on e-commerce stores. Most themes include partial Product schema but skip critical fields like AggregateRating, brand attribution, or proper availability state.
• Review and AggregateRating nested within Product schema to surface customer feedback in a verifiable format. This pairs with on-site review widgets or review apps. AI engines and Google cross-reference these aggregate ratings against third-party platforms like Trustpilot to validate authenticity.
• Offer nested within Product to handle pricing, currency, availability, shipping details, and conditions. For variant products, use individual Offer blocks per SKU rather than collapsing them.
• FAQPage on product and category pages where you have meaningful question-and-answer content. AI engines lean heavily on FAQ schema when answering "What is the best..." or "How do I choose..." queries.
• CollectionPage for category pages with curated product lists, especially for high-traffic shop pages.

One Shopify client we worked with started with only two schema types implemented by their theme. After we added the missing five (Product completeness, Review, Offer, FAQPage, BreadcrumbList) and validated them, they jumped from an AI visibility score of 31 to 84 in three months. Their Google search impressions also rose 155 percent, partly because rich snippets for products started appearing in search results. You can see more before-and-after stories on the FlinnSchema results page.

For more detail on the specific e-commerce considerations, see our AI visibility FAQs for e-commerce brands.

Local Service Businesses (Solicitors, Accountants, Trades, Salons)

If your business serves customers in a defined geographic area, the schemas that matter most are:

• LocalBusiness as your core entity schema. There are subtypes for many business categories (HairSalon, LegalService, AccountingService, Plumber, Dentist, and so on). Use the most specific subtype that applies because AI engines weight more specific types higher.
• Service blocks for each distinct service you offer, with name, description, area served, and ideally pricing or pricing range. One Service block per service type works better than collapsing everything into a single description.
• Review and AggregateRating nested within LocalBusiness. Pair with Google Business Profile and Trustpilot for cross-validation.
• FAQPage covering the questions real customers ask before engaging. AI engines pull from these directly when answering local recommendation queries.
• OpeningHoursSpecification if your hours vary or include special periods. Static daily hours can be expressed simply in LocalBusiness, but variable hours need the dedicated specification.

One of our long-running clients, a recruitment agency in Kent, had none of these implemented when we audited them. Their AI visibility score was 18 out of 100. After we added Organisation, LocalBusiness, Service blocks for each placement category, Review schema, and FAQPage, their score rose to 62 and they began appearing in answers from all four major AI engines.

If you serve a wider area than your physical address suggests, set the areaServed property explicitly. AI engines will not infer a 25-mile service radius from your office postcode. For more detail, our FAQs for local service businesses covers the common questions.

B2B Services, SaaS, and Consultancies

If you sell to other businesses rather than direct consumers, the schema mix is slightly different:

• Organisation with the most relevant subtype (ProfessionalService, ConsultingService, or SoftwareApplication for SaaS). Include detailed knowsAbout and areaServed properties because B2B AI queries often filter by specialism.
• Service blocks for each offering. For consultancies, one Service per practice area. For SaaS, you may use SoftwareApplication or Product per tier.
• FAQPage is especially valuable for B2B because buyers spend more time researching and AI engines surface FAQ content prominently in B2B queries.
• Person for named founders, principals, or notable employees. B2B trust signals often hinge on the credibility of the individuals behind the business. Person schema with sameAs links to LinkedIn and other profiles strengthens entity recognition.
• Article or BlogPosting on thought-leadership content with proper author attribution back to the Person schema.

FlinnSchema itself uses this pattern. Our root layout includes Organisation, Person for me as founder, multiple Service entries, and a WebSite block with search action. This is the same schema you can inspect on our homepage if you want to see a working example.

Content Publishers, Media, and Blogs

If your business model centres on publishing content for traffic and ad revenue (or to support a related business), the schemas you need are:

• Article or BlogPosting on every editorial page with headline, datePublished, dateModified, author, and image.
• Person for every named author, with proper sameAs links and credentials. This is critical for E-E-A-T signals that AI engines and Google use to assess content credibility.
• NewsArticle for news content specifically, if you publish time-sensitive stories. This unlocks placement in Google News and similar AI news features.
• VideoObject for embedded video content with thumbnail, duration, and transcript where available.
• HowTo for step-by-step instructional content, which AI engines surface heavily for "how do I" queries.

If your content reaches a niche audience, also consider schemas for the specific subject matter. Adventure tourism publishers might add Trip or TouristAttraction. Food and drink publishers might add Recipe (for actual recipes) or Brewery. The principle is to match schema to the actual content type rather than over-claiming generic schemas.

Multi-Location and Franchise Businesses

If you operate multiple locations under a single brand, the structure is:

• Organisation as the parent brand at the corporate level
• LocalBusiness for each individual location with its own address, phone, hours, and reviews
• Each LocalBusiness references the parent Organisation via parentOrganization
• Local pages should have the full schema for that location only, not the entire chain

This pattern signals to AI engines that you are a single trusted brand operating in multiple areas, rather than a collection of unrelated businesses sharing a name. The distinction matters when AI engines weigh credibility because the brand-level signals (history, reviews, mentions) attach to all locations.

Schemas You Can Probably Skip

Some schema types get a lot of attention but rarely help AI visibility for normal businesses. These include:

• Event unless you actually run events as a business function
• Recipe unless you publish recipe content specifically
• JobPosting only matters if your site is primarily a jobs board or you publish open roles regularly
• Course only for education businesses with structured courses
• SpecialAnnouncement was useful during COVID but has minimal value now

Adding these without genuine content underneath does not help your AI visibility and can hurt validation if you populate them with weak data. Better to skip them entirely. The same applies to overly speculative additions like CreativeWork blocks for products that are not actually creative works, or Brand schema implemented as a separate entity from your Organisation. Keep your schema honest and tightly tied to real content.

How to Actually Implement Schema

The technical format is JSON-LD embedded in your HTML, typically in the head or before the closing body tag. The official Schema.org documentation has the canonical reference for each type, and Google's structured data documentation covers Google's specific requirements where they differ.

For Shopify and WordPress sites, the practical implementation usually goes one of three ways:

1. Theme-level: edit your theme to inject the schema you need. This is the most flexible but requires developer time and re-implementation if you switch themes.
2. App or plugin-level: install a schema app or plugin that injects schema dynamically. This is faster but quality varies wildly across products, and many leave critical fields blank.
3. Platform integration: FlinnSchema's Shopify and WordPress integrations inject complete, validated schema automatically based on your store's actual data. This is the path we recommend for clients who do not want to manage schema manually.

Whichever path you choose, validate your output. Google's Rich Results Test and the Schema.org validator both flag issues. Test before you publish, then re-test monthly because plugins and themes can drift.

The Quickest Way to Know What You Need

If you would rather not work through the decision tree yourself, our free 26-factor audit identifies exactly which schema types your site is missing and ranks them by impact on your AI visibility score. The audit takes about 60 seconds and gives you a precise list of what to add, in priority order. For a deeper look at how the audit itself works, see inside the AI visibility audit.

For implementation support, our Premium plan covers the schema work end-to-end, including theme edits, validation, and ongoing monitoring. Or if you would rather talk through your specific business before deciding, book a free 15-minute walkthrough and we will look at your site live and tell you the schemas that would have the biggest impact.

For the deeper context on why schema matters so much to AI engines specifically, see how AI search engines decide which businesses to recommend, how to get cited by ChatGPT, and our broader piece on how AI visibility differs from SEO.