If you've heard the phrase "AI search" and quietly wondered what it means for your business, this is a good place to start. And if you've come across schema markup or structured data but assumed it was something for your developer to worry about — this post will give you enough to have an informed conversation about it.
Structured data for AI search is one of several useful things you can do to improve how your website is understood by search engines and AI tools. It's not a silver bullet, and it won't transform your visibility overnight on its own. But as part of a sensible SEO strategy, it's worth understanding and worth doing properly.
Here's what it is, why it matters, and where to focus your effort.
What Is Structured Data and Why Does It Exist?
Structured data is code added to your website — written in a format called JSON-LD, using a shared vocabulary from schema.org — that labels your content for machines. It doesn't change anything your visitors see on the page. What it does is give search engines and AI systems a clearer picture of what your content means.
Without it, a search engine has to interpret your website like a first-time reader with no prior knowledge. It can make reasonable inferences, and often does. But structured data removes the guesswork — and the clearer you make things, the better your chances of being understood accurately.
The schema.org vocabulary was created in 2011 as a collaboration between Google, Microsoft, Yahoo, and Yandex. Google has confirmed that JSON-LD is its preferred format, and it remains the most widely used implementation method today.
Despite being around for over a decade, structured data is still implemented properly by very few websites — which makes it a practical opportunity rather than a late-stage optimisation.
How It Relates to AI Search
Traditional search was built around keywords and links. You wrote good content, earned backlinks, and aimed to rank on a results page. That approach still matters — and still works.
AI-powered search adds a new layer. Tools like Google's AI Overviews, ChatGPT, and Perplexity don't just return a list of links. They generate answers — synthesised from multiple sources — and sometimes cite the pages they drew from. This is commonly referred to as Generative Engine Optimisation (GEO).
To be clear: appearing in AI-generated answers isn't guaranteed by any one tactic, including structured data. Content quality, authority, and topical relevance all play a significant role. What structured data does is reduce ambiguity — it helps AI systems understand what your business is and what you offer, which is a necessary (if not sufficient) condition for being cited.
Think of it as one strong signal among many, rather than a shortcut to visibility.
The Shift from Keywords to Entities
One of the useful concepts behind structured data is the idea of entities — clearly defined things like a business, a service, a location, or a person. AI systems increasingly organise information around entities and the relationships between them, rather than matching keyword strings.
When your website uses schema markup to say: “this is a business, this is what it does, this is where it operates, these are its services” — you're defining your entities in a way machines can work with confidently. The more complete and consistent that definition is across your site, the easier it is for search engines and AI tools to accurately categorise and surface you.
The Schema Types Worth Prioritising
Not every schema type is equally useful. Focus on the ones most relevant to your business type and content. Here are the most impactful for most businesses:
Organisation or LocalBusiness
This is your foundation. If you do nothing else, get this right. It tells search engines who you are and gives them the context to understand everything else on your site. For a local or service-based business, `LocalBusiness` is the right type to use.
Key properties to include:
- `name`
- `url`
- `telephone` and `email`
- `address`
- `areaServed`
- `description`
- `sameAs`
- `openingHours`
The `sameAs` property is worth particular attention. It links your schema to your presence on other verified platforms, helping search engines confirm your identity and establish credibility across the web.
Service
If you offer specific services, define each one using `Service` schema.
Include:
- `serviceType`
- `description`
- `provider` (linked back to your `Organization`)
Specificity matters here. Vague service descriptions add little value for either search engines or AI tools.
FAQPage
FAQ schema works well when the content is genuinely useful.
Guidelines:
- Each answer should stand on its own
- Aim for roughly 40–60 words per answer
- Avoid marketing copy disguised as FAQs
- Write in plain, conversational language
Article and BlogPosting
For content pages, include:
- `headline`
- `description`
- `datePublished`
- `dateModified`
- `author` (linked back to your organisation)
Keeping `dateModified` accurate is a simple signal that your content is being maintained.
Review
If you have genuine customer reviews, mark them up. Tying verified reviews back to your organisation adds credibility. Fabricated or incentivised reviews are not worth the risk.
BreadcrumbList
`BreadcrumbList` clarifies your site structure and page hierarchy. It’s a small implementation with a consistent payoff and should be deployed site-wide.
Building Schema as a System
Individual schema implementations help. A connected system helps more.
- Use the `@graph` pattern to define entities and relationships
- Keep `@id` values stable and consistent
- Only mark up content that is accurate and visible on the page
- Validate implementations using Google’s Rich Results Test and Search Console
What to Realistically Expect
Structured data is worth doing, but expectations matter:
- While structured data won’t immediately or directly improve rankings, it can increase visibility through rich results and AI chat responses.
- Increased visibility can drive higher engagement, which may influence SEO performance over time.
- It won’t compensate for weak content or low authority.
- It reduces ambiguity for both traditional search engines and AI systems.
- It’s one signal among many — not a replacement for solid fundamentals.
Conclusion
Structured data for AI search is a practical investment — not because it guarantees visibility, but because it makes your website clearer and easier for machines to trust.
If competitors haven’t implemented it properly, it’s an opportunity. If they have, it’s a baseline requirement. Either way, it’s not complex to get started, and the benefits compound when applied consistently.
If you'd like help putting a structured data strategy together, *[link to: Untapped contact page]* — we’ll give you a straight answer about what’s worth doing and what isn’t.
FAQs
What is structured data and why does it matter for my business website?
Structured data is code that labels your content for machines. It tells search engines and AI tools what your business is, what it offers, and where it operates, without ambiguity.
What is AI search optimisation and how is it different from traditional SEO?
Traditional SEO focuses on ranking in search results. AI search optimisation (often called GEO) focuses on being cited in AI-generated answers. Both matter, and structured data supports both.
Will adding schema markup improve my Google rankings?
Not directly. Structured data can improve visibility in rich results and AI answers, which can increase engagement and influence SEO outcomes over time.
Where should a small business start with structured data?
Start with `LocalBusiness` schema on the homepage. Get the basics right before expanding to services and FAQs.
How do I know if my structured data is working?
Use Google Search Console and the Rich Results Test. For AI search, manually check whether your business appears in AI-generated answers.
Do I need a developer to implement structured data?
Not always. Basic schema can be created using generators or CMS plugins, but a connected, site-wide approach benefits from technical expertise.
Any thoughts?
Leave a comment