Your competitor's Google result shows their star rating, three of their most-asked questions answered right on the results page, and a list of their services. Yours is a blue link and a URL.
That gap has a name. It is schema markup. And closing it does not require a developer, a technical SEO retainer, or an afternoon reading documentation written for people with computer science degrees.
This guide covers the five schema types that matter for service businesses, what each one does, and how to add them without touching code.
What schema markup actually does (and why it matters more in 2026)
Schema markup is structured data added to your website that tells search engines, in plain terms, what your business is, what it offers, and where it operates. One sentence. That is the definition.
The longer version: Google does not read your website the way a human does. It builds a model of your business from every signal it can find, including your page text, your backlinks, your Google Business Profile, and directory listings.
When those signals conflict or leave gaps, Google fills in the blanks itself. Schema replaces that guesswork with direct, structured answers.
This connects to entity SEO. Google categorises businesses as named entities, not just websites. Schema is how you confirm yours, telling Google that "City Roofing Ltd at 14 Trade Street, Birmingham" and the Checkatrade listing and the Google Business Profile are all the same business. That clarity feeds the knowledge graph and makes your business more legible across the entire Google ecosystem.
That legibility compounds.
In 2026 there is another dimension. When someone asks ChatGPT or Perplexity which physiotherapist to use in Birmingham, the AI is pulling from structured, legible data sources. Because AI search traffic converts differently to traditional clicks, there is a real business case for being findable by AI tools, not just by Google.
If your schema is absent or contradictory, AI systems have less to work with when constructing an answer about your business.
One thing schema does not do: directly improve your organic ranking position. Google has said so clearly. What it does is make your business clearer to Google, which improves click-through rates via rich results, increases your chances of appearing in AI responses, and compounds into better entity clarity over time.
Worth the hour it takes to implement.
Schema type 1: Organisation schema
Organisation schema is the foundation layer. It tells Google that your business is a real, named entity, distinct from every other business with a similar name.
It belongs on every page of your website, not just the homepage. Most WordPress sites add it to the site header so it loads sitewide automatically, which is the right approach.
The core properties are your business name, URL, and logo (added as an ImageObject). The most important property by some distance is sameAs. This is where you list every credible online profile that refers to the same business: your Google Business Profile, LinkedIn company page, Facebook page, and relevant directory listings.
For a physiotherapy practice, that sameAs array might include their GBP listing, their Physio First directory entry, and Companies House. For a firm of solicitors, it might include the Law Society's Find a Solicitor page, a legal directory, and LinkedIn. For a roofing contractor, Checkatrade, TrustATrader, and Companies House.
Get it right and entity clarity improves across every search context, including AI.
Why does sameAs matter so much? Google sees your business name mentioned in dozens of places across the web. The sameAs array is what tells it those mentions all point to the same entity. Here is the key structure:
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Midlands Physio",
"url": "https://www.midlandsphysio.co.uk",
"logo": {
"@type": "ImageObject",
"url": "https://www.midlandsphysio.co.uk/logo.png"
},
"sameAs": [
"https://www.google.com/maps?cid=YOURGMBID",
"https://www.linkedin.com/company/midlands-physio",
"https://www.physiofirst.org.uk/find-a-physio/midlands-physio"
]
}
Replace the placeholder URLs with your real profiles. The list does not need to be exhaustive. It does need to be accurate.
Schema type 2: LocalBusiness schema (and why the subtype matters)
LocalBusiness schema is the most impactful type for any business that serves customers in a specific geographic area. It feeds directly into local search results and the Google Maps pack.
The subtype is where most guides gloss over something significant. Rather than using the generic LocalBusiness type, you should use the specific subtype that describes your business. Google's knowledge graph is more precise when you give it a more specific entity class, and the right subtype connects your business to the correct category in search results.
For trades businesses: Plumber, Electrician, RoofingContractor, Painter, GeneralContractor. For healthcare: MedicalBusiness, Dentist, Physician, Optician. For professional services: LegalService, AccountingService, FinancialService.
These are all valid Schema.org types. Each one signals something meaningful to Google about the nature of your business.
Service-area businesses should use the areaServed property to list the towns and postcodes they cover. A roofer covering the West Midlands, a mobile physio visiting patients at home, a solicitor with offices across multiple cities. areaServed tells Google which locations to associate you with in local results, even where you have no physical premises.
One consistency rule that catches a lot of businesses out: the address and telephone number in your LocalBusiness schema must match exactly what is on your Google Business Profile and on your visible website. Mismatches create conflicting signals and Google trusts none of them.
Different phone number formats, different address layouts. Any variation counts. Check all three before you publish.
Here is a JSON-LD example for a roofing contractor, with multiple towns in areaServed:
{
"@context": "https://schema.org",
"@type": "RoofingContractor",
"name": "City Roofing Ltd",
"url": "https://www.cityroofing.co.uk",
"telephone": "0121 000 0000",
"address": {
"@type": "PostalAddress",
"streetAddress": "14 Trade Street",
"addressLocality": "Birmingham",
"addressRegion": "West Midlands",
"postalCode": "B1 1AA",
"addressCountry": "GB"
},
"areaServed": ["Birmingham", "Solihull", "Coventry", "Wolverhampton"],
"openingHours": "Mo-Fr 08:00-17:00",
"priceRange": "££"
}
Google's structured data documentation for LocalBusiness covers all required and recommended fields in one place. More readable than you might expect.
Combining LocalBusiness schema with a structured local SEO programme gives it the authority context it performs best in. Schema tells Google what your business is. A proper local search programme tells Google why it should trust you in a given area.
The two work together.
Schema type 3: Service schema
Most service businesses implement LocalBusiness schema and stop there. Service schema is the step almost nobody takes.
LocalBusiness schema tells Google you are a roofer. Service schema tells Google that you offer emergency roof repairs, new roof installations, and flat roof surveys as distinct, named, searchable services. Each with its own URL, its own description, and a link to a specific page on your website.
That distinction matters more than most people realise.
For a law firm, that means separate Service blocks for "Wills and Probate", "Commercial Leases", and "Employment Disputes", each pointing to the relevant service page. This maps directly to how a professional services firm already structures its website and navigation. No new content required.
For a physiotherapy practice, three Service blocks might cover an initial assessment, ongoing treatment, and sports rehabilitation. Google understands the practice as a provider of distinct, named, bookable services rather than a vague "physio place". That distinction matters in how it represents the business in AI responses and rich results.
The key properties for each Service are: name, serviceType, provider (linking back to your LocalBusiness entity), areaServed, description, and url. Schema.org's Service type documents all available properties. For most service businesses, those six are the ones that matter.
Healthcare note. For clinical procedures performed in a formal medical setting, Schema.org has a MedicalProcedure subtype used by NHS databases and clinical registries. For private practices, aesthetic clinics, and most healthcare service providers, the standard Service type is simpler, safer, and equally effective. Use MedicalProcedure only if your site operates in a formal clinical data context.
Schema type 4: FAQPage schema
FAQPage schema is the single type most directly connected to how AI Overviews are constructed.
When Google builds an AI Overview for a query like "do roofing contractors in Birmingham need planning permission?", FAQPage-marked content on relevant service business websites is among the sources it draws from. When a voice assistant answers whether a physio practice does home visits, the answer frequently comes from FAQPage-marked content on the practice's website.
This is also why it connects so naturally to answer engine optimisation. The goal is to make your content usable as a direct answer, not just as a ranked result. FAQPage schema makes that structure explicit.
Getting the questions right is what separates FAQPage schema that earns features from FAQPage schema that sits there doing nothing. "What services do you offer?" is too generic. "Do you cover emergency roof repairs in Birmingham at weekends?" is specific, location-relevant, and exactly what a real searcher types.
The more closely a question mirrors how a customer would actually phrase it, the more useful it is to both humans and AI systems.
Each answer should run 60-100 words. Substantive enough to be genuinely useful as a standalone answer, short enough to be surfaced in a voice result or AI Overview without being cut off. That window is tighter than most people think.
There is one rule Google is clear about: only add FAQPage schema to pages where the questions and answers are visibly displayed. Markup for hidden or non-existent content violates Google's guidelines and can result in the schema being ignored entirely. Do not add it to the header and hope nobody notices.
WordPress implementation. Rank Math (free tier) and Yoast SEO Premium both support FAQPage schema through their FAQ block editors. Add the FAQ section visibly on the page first, then enable schema for it in the plugin settings. No JSON-LD required.
Healthcare note. Questions like "Do I need a GP referral to see a physiotherapist?" are appropriate and useful. Questions that edge toward medical advice, "How should I treat a torn ligament at home?", should be answered carefully and with visible professional caveats on the page.
For businesses with AI search visibility in their sights, FAQPage schema is the single highest-leverage addition they can make to an existing site today.
That is a strong claim. It holds up.
Schema type 5: Review and AggregateRating schema
Review schema produces the visible win most people are hoping for when they first hear about schema markup: star ratings appearing directly in Google search results. A listing with a 4.8-star rating and 47 reviews next to it gets clicked more than a listing without.
Measurably so.
There is a condition Google rarely explains plainly, and it trips up a lot of businesses. AggregateRating schema only works on pages where the reviews are physically displayed on your website. You cannot point to your Google Business Profile reviews, your Trustpilot page, or any other third-party platform and use them as the basis for schema on your own site.
The reviews must live on your pages, visible to any reader.
The practical path: if the business already has a testimonials page or a reviews section on individual service pages where real customer reviews are displayed, AggregateRating schema can be added to show the aggregate score. If not, build that section first, then add the schema.
The required properties are ratingValue, reviewCount, and bestRating (typically 5). The values must be accurate. Google checks these against the displayed reviews. Inflated or estimated counts are what cause the feature to be ignored, or in serious cases, a manual action from Google's quality team.
Healthcare note. Aggregate ratings based on patient experience of service quality, communication, and the booking process are generally appropriate. Ratings framed around clinical outcomes or the effectiveness of a specific treatment enter YMYL territory and carry more risk. Keep review content focused on the care experience rather than the clinical result.
Review schema is listed last not because it matters least, but because it requires the most infrastructure to do correctly. The on-site reviews section must exist before the schema has anything to work with.
Build the foundation before the feature.
How to add schema markup without writing a single line of code
There are three realistic routes for a service business without a developer on call.
1. WordPress plugin.
Rank Math's free tier covers Organisation, LocalBusiness, FAQ, and Review schema through a guided settings interface. You fill in plain text fields and the plugin generates the correct JSON-LD. Yoast SEO Premium offers a polished FAQ block editor and is included in the premium plan if the site already uses it. For most WordPress sites, this is the right starting point.
2. Schema generator, then paste.
Tools like TechnicalSEO.com's schema generator produce valid JSON-LD from a form you fill in with plain English. The output is a script block you paste into the custom header injection field of your website builder. Squarespace, Wix, and Webflow all have one. No coding required beyond copy and paste.
3. Manual JSON-LD.
For businesses with a developer available, pasting a JSON-LD script block directly into the <head> of the site template gives the most precise control. The examples in this guide show the format. Straightforward once you have done it once.
Testing. Validate every schema implementation with Google's Rich Results Test before the page is indexed. Paste the URL or the code block, and the tool shows you immediately whether the schema is valid, which properties were detected, and whether any required fields are missing. Fix errors before moving on.
Do not skip this step.
Implementation sequence. Do not try to add every type at once. Start with Organisation schema sitewide. Then LocalBusiness on your homepage and contact page. Then Service schema on each individual service page. Then FAQPage on pages that already have visible FAQ sections. Then AggregateRating once on-site reviews are in place. Each step builds on the last, and the sequence matters.
Schema is necessary groundwork. But it is not the whole answer.
Schema tells Google what your business is, what it offers, and where it operates. Done well, it enables rich results, improves click-through rates, and makes your business more legible to AI systems.
That much it does reliably.
Whether ChatGPT, Claude, and Perplexity are actively recommending your business to people searching right now, that is a separate question. Schema alone does not answer it.
If you want to know where you currently stand in AI search, the AI Visibility Audit will tell you in one report. It checks whether the major AI platforms can accurately represent your business, what they are actually saying about you, and where the gaps are.
Free resource: AI Visibility Audit
Find out whether ChatGPT, Claude, and Perplexity are representing your business accurately (or at all) in one report, covering your current AI search visibility and where the gaps are.
Or if you would rather have someone handle the implementation and the broader search strategy on your behalf, technical SEO work is part of every programme our SEO agency runs for service business clients.
Frequently asked questions about schema markup for service businesses
Does schema markup directly improve my Google ranking?
No, and Google has stated this clearly. Schema markup is not a ranking factor in the way that links and content quality are. What it does is make your business more legible to Google, which enables rich results such as star ratings and FAQ panels that improve click-through rates from search.
It also contributes to entity clarity over time. The honest case for schema is not "it will rank you higher". It is "it will help what you rank for perform better in the results it already appears in".
What is the difference between JSON-LD and microdata?
JSON-LD is a small structured code block added to the <head> of a web page. It describes your business, its properties, and its relationships without touching any visible page content. Microdata is an older approach that embeds structured data directly inside the visible HTML of the page, requiring you to annotate every element.
Google recommends JSON-LD for all new implementations because it is cleaner, easier to maintain, and does not require changes to visible page content. If an older website has microdata already in place, it still works. But JSON-LD is the format to use for anything new.
Which schema type should I add first?
Organisation schema. It is sitewide, takes the least time to set up, and is the foundation everything else builds on. Once Organisation schema is in place, add LocalBusiness to your homepage and contact page, then Service schema to each service page, then FAQPage to pages with visible FAQ sections, then AggregateRating once you have on-site reviews to support it.
That sequence matters. Skipping to AggregateRating without the foundation layers in place is a bit like painting a wall before you have plastered it.
Does schema markup help with AI Overviews?
Yes, particularly FAQPage schema. When Google constructs an AI Overview for a query like "what does a boiler service include?" or "do I need planning permission for a loft conversion in the UK?", it frequently draws from FAQPage-marked content on relevant service business websites.
The more specific and clearly written the Q&A content, the better the chance of being surfaced. Schema does not guarantee an AI Overview placement. What it does is give Google the structured, citable content it needs to include your business in AI-generated answers, rather than having to infer it from unstructured text.
