Schema markup is one of those quiet, invisible parts of a website that most owners have never asked us to add by name, and yet in 2026 it has become one of the most important decisions a small business will make about its website this year. Google's AI Overviews now answer more than half of all UK searches before the user has scrolled. ChatGPT, Claude and Perplexity quietly field shopping and service queries inside the chat itself. Google's May 2026 core update has tightened the link between structured data and visibility further still. Without a proper schema layer, a perfectly written website can sit invisible to the AI systems that are now intermediating a fast growing share of customer research, and the cost of that invisibility is real.
This guide is for the Suffolk and Cambridgeshire business owner who has heard the words 'schema markup' once or twice from a developer or an SEO agency, has nodded politely, and has not really known what to do about it. We will cover what schema markup actually is in plain English, why 2026 has made it essential rather than optional, which schema types matter most for a small business website, how to test whether your site already has any, the common mistakes that turn a well meant implementation into a quiet ranking risk, and what FutureProofs ships as standard in every Webflow build at the £1,995 Business package level.
Schema markup is a small invisible layer of structured information that sits inside the code of a webpage and tells a search engine, an AI Overview or a ChatGPT response in unambiguous terms what the page is actually about. A regular paragraph on a website might mention that the company is called Riverside Plumbing, that it serves Bury St Edmunds and Newmarket, that the phone number is 01284 123456, that it opens at eight in the morning, that the average review is 4.9 stars across 137 Google reviews, and that the founder is Gas Safe registered. A search engine has to interpret all of that from prose, and prose is, by its nature, ambiguous. Schema markup says the same thing in a form designed for machines: this is a LocalBusiness, its name is Riverside Plumbing, its address is X, its phone is Y, it has 4.9 stars from 137 reviews, it is Gas Safe registered with number Z, all wrapped in a small JSON-LD block that sits politely in the page's source code and does not affect how the page looks to a human visitor at all.
For most of the last decade, schema markup was sold to small businesses as a way to earn rich snippets in Google: the star ratings, the FAQ accordions, the recipe cards. In 2026 the picture has changed completely. AI Overviews now sit on top of more than half of UK Google searches, and they decide which businesses to cite by reading the structured data from candidate pages first and the prose second. ChatGPT, Claude and Perplexity all use the same kind of structured signals when deciding which UK service business to recommend in a conversational answer. Without schema markup, a website is asking an AI system to figure things out from context alone, and given that those systems are now answering an estimated half of all queries without sending the user to any website at all, the website that does not earn its way into the AI answer is invisible at exactly the moment that matters.
The May 2026 Google core update has accelerated this further by rewarding pages whose claims can be cleanly verified through structured data, which is why we now treat schema as a foundational layer rather than a polish. A website with proper schema is legible to the new web. A website without it gets quietly skipped.
For a small business website the schema landscape is simpler than the documentation makes it look. There are around half a dozen schema types that genuinely matter for a small business, and the rest can be ignored without consequence.
Every site should publish an Organization or LocalBusiness schema block on the home page covering name, logo, address, phone number, opening hours, geographic service area and social profiles. Service businesses should publish a Service schema block on each service page describing what is offered, who delivers it, and roughly what it costs. Ecommerce sites should ship Product schema with price, availability and reviews on every product page.
FAQPage schema on FAQ style sections, including the FAQ block at the bottom of every blog post, gives AI Overviews a clean structured source to lift from. BreadcrumbList schema makes navigation legible to search engines. Article schema with named author, publisher and publication date is the bare minimum for content like this very piece. Most Suffolk and Cambridgeshire small businesses need exactly that handful, no more, and none of them should be skipped on the assumption that 'we are not big enough to bother'.
There is a popular myth that schema markup is read once a year by a bot and quietly forgotten. The reality in 2026 is the opposite. Every time an AI Overview is composed, every time a Perplexity user asks 'best accountant in Cambridge for sole traders', every time ChatGPT is asked about a hairdresser in Bury St Edmunds, the answer is composed in real time from structured signals across dozens of candidate pages. The systems weight clean, well formed schema very heavily because it removes ambiguity. A page that says 'we serve Cambridge and surrounding areas' in prose is much weaker than a page that has the same statement in prose plus a Service schema block listing 'areaServed: Cambridge, Newmarket, Ely, Huntingdon' in machine readable form.
For a Suffolk business, the consequence is that without schema you are competing for AI citations with one hand tied behind your back, regardless of how good your prose, your photography or your reviews actually are. The website that publishes its services, areas and reviews in a structured form is the website that gets quoted.
Most existing UK small business websites, particularly older WordPress builds running thinly maintained theme plugins, either have no schema markup at all, or have a partial, incorrectly nested implementation that Google's testing tools mark as invalid. The fastest way to check is to drop your website's URL into Google's Rich Results Test at search.google.com/test/rich-results and Schema.org's own validator at validator.schema.org. Both tools take less than thirty seconds and will tell you which schema types they detect, which fields are missing, and where the errors are.
If the response comes back showing 'no items detected', or a long list of warnings about missing required properties, the site has effectively no useful schema layer. We run this test on every Suffolk and Cambridgeshire client during the FutureProofs onboarding audit, and roughly four out of every five existing small business websites we look at fail it on first inspection. It is the single most common technical gap we find.
Every Webflow site we ship at the £1,995 Business package level includes a complete AI SEO schema layer as standard, not as an upsell or an add on. We configure Organization or LocalBusiness schema on the home page with full address, phone, opening hours and geographic service area. We add Service schema on each service page. We add FAQPage schema on the FAQ blocks at the bottom of every blog post. We add BreadcrumbList schema on every navigable page. We add Article schema on every blog post with named author, publisher and publication date.
The Starter package at £895 includes the home page LocalBusiness schema as a minimum, which is enough to put a small site on the map. Our SEO Foundation retainer at £495 per month maintains and extends the schema as the business grows, with new locations, new services and new reviews added as they happen, and the website management plan at £95 per month keeps the technical layer clean as Google's structured data guidelines change. The Ecommerce package at £3,995 adds Product, Offer and Review schema across every product page. All pricing is on futureproofs.co.uk/pricing/ in plain numbers.
The two most common mistakes we see when small businesses attempt schema markup themselves, or accept it from a generalist developer who does not specialise in SEO, are the same year after year. The first is marking up information that is not actually visible on the page. The second is marking up reviews that the business has not actually received. Both are explicitly against Google's structured data guidelines, both can trigger manual penalties, and both happen surprisingly often when an agency is trying to inflate a thin website with synthetic schema in the hope of unlocking rich snippets.
The other common mistake is the opposite: adding schema only to the home page and forgetting that the service pages, the contact page and the blog posts each need their own appropriate schema type. A clean schema implementation in 2026 is honest, page by page, only describes what is genuinely on the page, and is validated through Google's Rich Results Test before going live. Anything else is wasted work at best, and a real ranking risk at worst.
What is JSON-LD and is it the same as schema markup?
JSON-LD is the format used to write schema markup, and in 2026 it is the only format Google, Bing and the major AI search systems genuinely recommend. The older Microdata and RDFa approaches have been quietly retired in practice. JSON-LD is a small structured block of code, usually placed in the head of a page, that wraps schema.org definitions in a clean, machine readable form. Schema markup is the conceptual layer, what you are saying about the page. JSON-LD is the practical layer, how you are saying it. For any Suffolk small business getting schema added in 2026, the implementation should default to JSON-LD without exception.
Does schema markup directly improve Google rankings?
Schema markup does not directly raise a page's rank in the traditional sense. Google has been clear for years that it is not a ranking factor in the way that backlinks or content quality are. In 2026 it has become an indirect ranking lever of considerable power, because schema is what allows AI Overviews to cite you, what unlocks rich snippets like FAQ accordions and review stars in regular search results, and what feeds ChatGPT and Perplexity the structured signals they use to recommend businesses. Click through rates on results with rich snippets are typically twenty to thirty five per cent higher than those without, and AI citations are a separate, growing channel of visibility entirely.
How long does it take for schema markup to start working?
Schema markup is typically picked up by Google within a few days to a few weeks once a page has been recrawled. Rich snippets often appear in search results within four to eight weeks of a clean implementation going live. AI search systems like ChatGPT and Perplexity update their indexes on different cycles, generally faster, and citations from those systems can begin appearing within days. The slower moving piece is the AI Overview composition layer, which weights pages by overall trust and content quality as well as schema. For a brand new site with no review history, expect three to six months for schema to translate into consistent AI Overview visibility.
Do I need schema markup if I have a Google Business Profile?
A Google Business Profile and schema markup do related but quite different jobs, and a serious Suffolk small business needs both. The Business Profile feeds Google's local map pack, the three business map at the top of local searches, and is essential for being found on a 'plumber Bury St Edmunds' style query. Schema markup feeds Google's regular search results, AI Overviews and the AI search engines like ChatGPT and Perplexity. They share information, name, phone, address, hours, and Google strongly prefers the two to match exactly, but neither replaces the other. Treating one as a substitute for the other is a common and costly mistake.
How much does it cost to add schema markup to a website in the UK?
For a small UK business, adding schema markup as a one off retrofit to an existing website typically costs between £400 and £900 depending on the number of pages and the technical condition of the existing site. WordPress sites run into the higher end of that range because of theme and plugin conflicts. Adding schema as part of a fresh Webflow build is dramatically cheaper because it is built into the project from day one. At FutureProofs the £1,995 Business package includes the full AI SEO schema layer at no separate cost, and the £895 Starter package includes the home page LocalBusiness schema. A retrofit on an old WordPress site can sometimes cost more than a fresh Webflow build that ships with the schema already in place.
Can I add schema markup to a WordPress site myself?
You can technically add schema markup to a WordPress site yourself using plugins like Yoast SEO, Rank Math or Schema Pro, and for a basic LocalBusiness block on a home page that route is reasonable. In practice the result is rarely complete or correct without an experienced SEO hand on the configuration. The plugins generate generic templates that often miss properties like areaServed, priceRange and aggregateRating, and they frequently nest schema in ways that fail Google's Rich Results Test on validation. For a serious 2026 implementation across home, services, FAQ and blog pages, professional configuration is the safer route, and is what we charge in at FutureProofs.
Will schema markup help my site appear in ChatGPT and Perplexity?
Yes, schema markup demonstrably helps a website appear in ChatGPT, Claude, Perplexity and Google's AI Overviews. All of those systems rely on structured data as one of the highest quality signals when deciding which businesses to cite in a conversational answer. The mechanism is slightly different for each. ChatGPT and Claude consume a snapshot of the live web and weight schema rich pages heavily. Perplexity reads pages live during each query and prioritises sources with clean structured data. Google's AI Overviews compose answers from a real time mix of structured and unstructured signals. The common thread is that schema poor sites systematically lose visibility in all of them.
What is the difference between LocalBusiness, Organization and Service schema?
Organization schema is the broadest description of a company and is appropriate for any business with a website, regardless of whether customers visit a physical location. LocalBusiness schema is a more specific subtype that adds geographic and opening hours properties and is the right choice for any Suffolk or Cambridgeshire business with a service area or a premises, including accountants, plumbers, restaurants and clinics. Service schema describes a particular service the business offers, with its own audience, area served and pricing, and is layered on top of LocalBusiness on each service page. In a typical FutureProofs Business build, all three coexist: LocalBusiness on the home page, Service schema on each service page, Article schema on the blog.
Schema markup is no longer a polish layer applied to a finished website. It is a foundational structural decision that determines whether a Suffolk small business is visible in the AI search systems that now intermediate more than half of all UK Google queries. Get it right at the start of the build, in JSON-LD, on every relevant page type, validated through Google's Rich Results Test, and the website will be cited fluently across ChatGPT, Perplexity, Claude and Google's AI Overviews for years to come. Get it wrong, or skip it, and the most expensive website in Bury St Edmunds will go quietly invisible at exactly the moment AI search reaches mainstream adoption. Pricing is on futureproofs.co.uk/pricing/.
