What schema is, the five types your site needs, and why structured data matters for AI search even after Google’s recent rich result updates.
Schema markup is structured code that tells search engines and AI tools exactly what your website is about, and many B2B sites have little or none. Here is what it does, the five types worth having, and why it still matters now that Google has retired its most popular rich result.
Table of Contents
- What is schema markup?
- The five schema types B2B websites need
- What happens without schema markup
- Schema types come and go: what the FAQ story teaches us
- How to check your schema markup (and who should add it)
- The five schema types all B2B websites should use
- Frequently asked questions about schema markup
What is schema markup?
Schema markup is a standardised vocabulary of code. It is maintained at Schema.org and founded by Google, Microsoft, Yahoo and Yandex. Schema states the facts of a web page outright: this is an organisation, this is its legal name, this is an article, this person wrote it, on this date, etc. It is added to a page as a small block of JSON-LD, a format invisible to human visitors and read directly by machine crawlers.
The problem it solves is inference. Without markup, a search engine or AI tool has to deduce what your business is and who wrote your content from prose alone, and deduction brings uncertainty. A machine that is uncertain about your facts is less likely to treat you as a distinct, verified entity, and less likely to cite you. We covered what those machine readers can and cannot see in our guide to how AI crawlers read your website; schema is the layer that removes the guesswork for your brand’s key facts.
The five schema types B2B websites need
The full Schema.org vocabulary runs to hundreds of types, which sounds daunting until you notice how few apply to you. Recipe, JobPosting, Product and MedicalWebPage exist for recipe sites, job boards, online shops and medical publishers. A typical B2B services website needs about five.
Organisation schema: the one to do first
Organisation schema states your legal name, URL, logo, contact details and official profiles, and its sameAs property points machines at the external records that corroborate you, such as your LinkedIn page and Companies House listing. It is the single type that establishes your brand as a verifiable entity, which is why we treat it as the first job on any site.
Article and BlogPosting schema
Applied to every post, this states the headline, author, publication date and date updated outright, feeding the freshness and authorship signals that search engines and AI tools both read.
Person schema
Person schema connects a named author to their credentials, profiles and your organisation. It is the structural half of the authorship work described in our guide to E-E-A-T, turning a byline into a verifiable claim.
FAQPage schema
FAQPage schema formats question-and-answer pairs for direct machine extraction. Its use took a sharp turn this year, covered below, but the format remains one of the most readily extractable structures a page can carry.
BreadcrumbList schema
BreadcrumbList states where a page sits within your site’s structure, helping machines understand how your topics relate to one another.
What happens without schema markup
A site without markup is not invisible; it is ambiguous. Machines must infer your facts, may not recognise your brand as a distinct entity, and cannot reliably connect your authors’ credentials to your organisation. Each of those uncertainties quietly lowers the odds of your content being cited without anything on your visible page looking out of place.
Schema is the most consistent gap in the free AI readiness checks we run. Well-designed B2B sites with genuinely useful content are let down by absent or generic markup, and teams who did not realise there was an issue.
Schema types come and go: what the FAQ story teaches us
Individual schema types get retired, and the most instructive recent example is FAQ. Google restricted FAQ rich results in August 2023 (along with HowTo results), announcing they would only be shown for well-known, authoritative government and health websites (Google, 2023). HowTo results began winding down, which were fully retired weeks later. In May 2026 the withdrawal completed: Search Engine Journal reported that FAQ rich results no longer appear in Google Search at all, with reporting and testing support removed over the months that followed (Search Engine Journal, 2026).
If schema were only a trick for winning rich snippets, that would be the end of FAQPage markup. But, fortunately, it is not for two reasons. Google’s own guidance says there is no need to remove existing FAQ structured data; it causes no harm sitting on the page (Google, 2023). And, importantly, Google Search was never the only reader: AI crawlers scan structured question-and-answer pairs whether or not a visual search feature rewards them, because the format hands them an extractable answer with its question attached. We keep FAQPage markup on our own articles for exactly this reason and we still recommend our clients use them too.
The wider lesson is about how to judge structured data. Any single type can be deprecated when a search feature is retired, so measure schema by what it communicates to machines and by extension other people, not by which platform’s display feature it currently earns. The facts you mark up outlast the features built on top of them.
Get your schema checked in one pass
Our free AI Readiness Check reviews your schema alongside crawler access, content structure and what AI tools currently say about your brand, in a single prioritised report.
How to check your schema markup (and who should add it)
Two free tools check your schema. Google’s Rich Results Test shows which Google-supported types a page carries and whether they validate, though its FAQ support was removed in mid-2026 along with the rich result itself. The Schema Markup Validator checks any Schema.org type, FAQ included, which makes it the better all-purpose check now. Run one or the other after any significant site change.
As for who does the work: adding JSON-LD is a paste job in most content management systems, through an SEO plugin or a custom HTML block, and the templates for the five types above are stable and reusable. You do not need a developer for a standard setup, though larger or custom-built sites may benefit from one. It is also among the first things we review and fix quickly following an AI readiness audit.
The five schema types all B2B websites should use
Organisation, Article, Person, FAQPage and BreadcrumbList: five stable templates that turn a site machines must guess about into one they can verify. The features built on schema will keep changing, but the facts you mark up will keep being read.
Learn more about how AI and machines read your website with our plain-English guide to AEO, and then why not request a free AI Readiness Check and we will show you exactly which types your site is missing.
Frequently asked questions about schema markup
The questions we get most from B2B marketing and web managers.
What is schema markup?
Schema markup is standardised code, usually in JSON-LD format, added to a web page to state its facts explicitly to machines: what the business is, who wrote the content, when it was published. It is invisible to human visitors and read by search engines and AI crawlers, removing their need to infer.
Does schema markup affect SEO?
Yes, indirectly and usefully. Schema does not raise rankings on its own, but it makes pages eligible for certain search features and, more usefully now, gives search engines and AI tools verified facts to work from. A machine that is certain about your entity and authorship is more likely to surface and cite you.
What is Organisation schema?
Organisation schema is the structured data type that states a business’s legal name, website, logo, contact details and official profiles, with a sameAs property linking to corroborating records such as LinkedIn and Companies House. It establishes the brand as a verifiable entity, and it is the first type any B2B website should add.
Is FAQ schema still worth using?
Yes. Google retired FAQ rich results in May 2026, so the markup no longer earns an expanded Google listing, but Google’s guidance confirms there is no need to remove it, and AI tools still read structured question-and-answer pairs when extracting answers. The visual feature is gone; the machine readability is not.
Do I need a developer to add schema markup?
Usually not. In most content management systems, JSON-LD schema is added through an SEO plugin or pasted into a custom HTML block, and the common B2B types follow stable, reusable templates. A developer is worth involving for large or unusual sites, or where markup needs generating automatically across many pages.