How to Appear in Google AI Overviews

Google AI Overviews — the AI-generated summaries that appear above organic search results for hundreds of millions of queries — have fundamentally changed what it means to be visible in Google.

Ranking on page one used to be the goal. Now you can rank first and still be invisible if the AI Overview answers the query without citing you.

Understanding how Overviews work is the first step to appearing in them.

How Google AI Overviews work

AI Overviews are generated by Google’s Gemini model, which synthesises information from multiple sources to produce a direct answer to a search query. The sources it draws from are not simply the top-ranking pages — they are selected based on a more complex set of signals.

The key factors Google’s documentation and research point to:

Entity strength in the Knowledge Graph. Google’s AI systems heavily weight sources and brands that have strong, coherent entities in the Knowledge Graph — the structured layer Google has built over the past decade that describes what people, places, businesses and concepts actually are.

Schema markup and structured data. Pages with clear, accurate schema markup that tells Google’s systems exactly what type of entity the page is about, what claims it makes, and in what context are significantly more likely to be extracted and cited.

E-E-A-T signals (Experience, Expertise, Authoritativeness, Trustworthiness). AI Overviews weight these signals more heavily than traditional rankings did. Authorship information, organisational credentials, corroborating mentions in authoritative sources — these are the trust signals the model uses to decide whether your content is worth surfacing.

Direct, extractable answers. Content structured to answer a specific question directly — particularly in the first paragraph, with supporting detail following — is more likely to be extracted than content that buries the answer in narrative.

What doesn’t work

Before covering what works, it is worth being clear about what doesn’t — because many of the instinctive responses to “how do I appear in AI Overviews?” are ineffective or counterproductive:

• More content. Publishing more articles at the same quality level will not improve your AI Overview presence. The model has more than enough text to work with. What it needs is better-structured, more authoritative text.
• More backlinks. Traditional link-building has almost no direct impact on AI Overview selection. Citation in the right authoritative sources matters; backlink count does not.
• Keyword optimisation. Writing for keyword frequency is actively counterproductive. AI systems are sophisticated enough to recognise intent and topic relevance without keyword stuffing, and over-optimised content reads as low-quality to the model.

What does work

1. Strengthen your Knowledge Graph entity

If your business has a weak or absent Knowledge Graph entry, improving it is your highest-leverage action. This means: claiming and completing your Google Business Profile with full attributes, ensuring your business is listed consistently across structured directories, building a Wikidata entry, and creating schema markup that connects your website clearly to your Knowledge Graph entity.

2. Implement comprehensive schema markup

At minimum, your website should have Organization or LocalBusiness schema with accurate, complete attributes. If you publish content, Article schema with correct author (connected to a Person entity with credentials) and datePublished fields matters. FAQ schema on pages that answer specific questions has historically been extracted for featured positions and remains relevant for AI Overviews.

3. Structure content for extraction

Write in a format AI systems can extract: direct answer in the first paragraph, supporting reasoning and evidence following. Use specific, attributable claims. Avoid passive constructions and marketing language that hedges every statement. The model needs to be able to attribute a clear answer to your page.

4. Build authoritative corroboration

Identify three to five sources that Google recognises as authoritative in your industry — trade press, accreditation bodies, structured directories, government sources where applicable — and build genuine presence there. The corroboration of your entity across authoritative sources is one of the strongest trust signals available.

5. Fix E-E-A-T signals

Make authorship explicit and credentialled. Ensure your About page clearly describes the expertise and experience behind your content. Link to external corroboration of your credentials. This is particularly important in YMYL (Your Money Your Life) categories — finance, health, legal — where AI Overviews are more conservative about which sources they cite.

The measurement question

One of the frustrations with AI Overview optimisation is that traditional rank-tracking tools do not measure AI Overview appearances. You need to test manually — searching queries you care about and recording whether and how your brand appears — or use tools specifically designed for generative engine monitoring.

Building a measurement baseline before you start work means you can actually see the improvement. Without measurement, you are optimising blind.

This, incidentally, is why the Entity Audit at geo.bz starts with establishing exactly this baseline.