How to get into Google AI Overviews: the complete 2026 guide
TL;DR
Google AI Overviews (formerly SGE – Search Generative Experience) appear in almost one in three searches in Slovakia in 2026. An online store featured in AI Overviews gets a direct recommendation before the organic results even start. The keys to appearing: structured content, schema.org markup, FAQ sections and a strong entity. This article shows you exactly what to do.
What Google AI Overviews are
Google AI Overviews is a feature Google launched globally in May 2024; it has been available in Slovakia and Czechia since September 2024. When you search, an AI-generated summary answering your question appears at the top of the page — above both the organic results and paid ads.
Unlike a classic Google results page with 10 blue links, AI Overviews answers you directly: "For 'where to buy a good-quality mattress in Slovakia', we recommend these online stores: …" and attaches 3–5 sources the AI relies on.
For online stores this has a major impact. A click from AI Overviews comes from a customer who is further along the buying journey — they have already received an AI recommendation and are now just checking the details. According to Search Engine Land analyses from 2025, the conversion rate of such visitors is up to 34 % higher than for a classic organic click.
How AI Overviews work technically
Understanding the mechanism helps you target your optimisation. Google AI Overviews uses the Gemini 1.5 Pro model, which:
- Identifies the intent of the search — what exactly the user wants to achieve
- Searches the web in real time (web grounding) for relevant sources
- Synthesises an answer from sources that meet its quality criteria
- Cites its sources — showing the 3–7 websites it drew information from
The criteria Gemini uses to pick sources for AI Overviews are similar to the E-E-A-T signals of classic SEO, but with an emphasis on content structure, an answered question and schema.org metadata.
Why AI Overviews "skip" online stores
An analysis of 500 Slovak online stores that we ran at Pozorio AI in January 2026 revealed 3 main reasons why stores are missing from AI Overviews:
| Problem | % of stores | Impact |
|---|---|---|
| Missing Product/FAQ schema | 78 % | The AI cannot recognise the page type |
| AI crawlers blocked in robots.txt (search crawlers such as OAI-SearchBot are the critical ones) | 34 % | ChatGPT and other AI search engines cannot see you (AI Overviews are unaffected — they are governed by Googlebot) |
| Generic descriptions with no added value | 91 % | The AI prefers another source |
The worst combination: no schema.org + robots.txt blocking crawlers + a description copied from the manufacturer. A page like that gives AI Overviews nothing to cite.
7 steps to get into AI Overviews
1. Unblock AI crawlers in robots.txt
This is the first step; without it nothing else works. Google Overviews uses Googlebot (the same one as classic SEO), but checking for blocks on AI-specific crawlers matters too.
Check your robots.txt:
# This MUST NOT be in robots.txt if you want to appear in AI Overviews:
User-agent: Googlebot
Disallow: /products/
# This is fine — it only blocks admin sections:
User-agent: *
Disallow: /admin/
Disallow: /checkout/
If you see User-agent: Googlebot with Disallow: rules for product pages, you have a serious problem. Fix it immediately.
2. Implement Product schema on every product page
Google Overviews loves structured data. Product schema with complete attributes is your ticket into AI recommendations.
The minimal version:
{
"@context": "https://schema.org",
"@type": "Product",
"name": "ErgoMax Pro ergonomic office chair",
"description": "Adjustable office chair with lumbar support, a breathable mesh seat and a 5-year warranty.",
"brand": {
"@type": "Brand",
"name": "ErgoMax"
},
"offers": {
"@type": "Offer",
"price": "349",
"priceCurrency": "EUR",
"availability": "https://schema.org/InStock",
"priceValidUntil": "2026-12-31",
"seller": {
"@type": "Organization",
"name": "YourShop.com"
}
},
"aggregateRating": {
"@type": "AggregateRating",
"ratingValue": "4.7",
"reviewCount": "143"
}
}
Key attributes that raise your chance of a citation: aggregateRating (rating and number of reviews), availability (stock availability), brand (brand identification).
3. Add FAQPage schema to category pages
Category pages (e.g. "Office chairs") are the ideal place for an FAQ section with FAQPage schema. Google Overviews cites them for questions such as "which office chair is best".
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "How do I choose the right office chair?",
"acceptedAnswer": {
"@type": "Answer",
"text": "When choosing an office chair, look at: seat height (adjustable 40–55 cm), lumbar support (adjustable to the lower back), seat material (mesh for breathability, fabric for warmth), load capacity (min. 120 kg) and warranty (at least 2 years)."
}
},
{
"@type": "Question",
"name": "How much does a good office chair cost?",
"acceptedAnswer": {
"@type": "Answer",
"text": "A quality office chair for 8 hours of sitting a day costs 150–500 EUR. Cheaper models under 100 EUR lack sufficient ergonomics for long-term sitting and usually last less than 3 years."
}
}
]
}
4. Write content that answers questions directly
AI Overviews are triggered by informational and comparison searches — that is, when a customer is still looking for information rather than a specific product. If you want to be cited, you need content that answers directly.
Doesn't work: "Our chairs are the best on the market."
Works: "An ergonomic chair must have adjustable lumbar support, a seat height of 40–55 cm and a load capacity of at least 110 kg. Without these parameters you risk health problems from prolonged sitting."
Specific, verifiable information + a clear answer to the question = an AI Overviews signal.
5. Optimise loading speed (Core Web Vitals)
Google AI Overviews only cites pages it would be willing to show in Google results — that is, pages with good Core Web Vitals. Minimum requirements for 2026:
- LCP (Largest Contentful Paint): < 2.5 seconds
- FID / INP (Interaction to Next Paint): < 200 ms
- CLS (Cumulative Layout Shift): < 0.1
Check on PageSpeed Insights — if your scores are red, this is a priority before any content optimisation.
6. Build your brand entity
For AI Overviews to know who you are, you need to be defined as an entity — a consistent identity recognisable across the web.
Practical steps:
- Fill in your Google Business Profile completely (category, description, photos, opening hours)
- Create or update a Wikidata entry for your company
- Keep NAP data (Name, Address, Phone) consistent on your website, Heureka and Firmy.sk
- Add Organization schema to your homepage with
sameAslinks to your social profiles
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "YourShop.com",
"url": "https://www.yourshop.com",
"sameAs": [
"https://www.facebook.com/yourshop",
"https://www.instagram.com/yourshop",
"https://www.heureka.sk/obchod/yourshop"
]
}
7. Monitor your appearances in AI Overviews
The problem with AI Overviews: Google Search Console does not report AI Overviews clicks separately. They are included in the regular organic statistics without being flagged.
To measure your real visibility in AI Overviews you need external monitoring — regular testing of whether AI Overviews cites your pages. That is exactly what Pozorio AI does automatically for your key queries.
When AI Overviews don't work for online stores
Not every search triggers AI Overviews. Google shows them primarily for:
- Informational queries: "how to choose a mattress"
- Comparison queries: "best ergonomic chairs"
- Local queries: "where to buy an ergonomic chair in Bratislava"
Less likely AI Overviews triggers:
- Navigational queries: "yourshop.com" (the user wants your website directly)
- Transactional queries: "buy ergonomic chair 349 EUR" (Google prefers to show a Shopping ad)
Focus on informational and comparison keywords when creating FAQ and category content.
Frequently asked questions about Google AI Overviews
Can I ask AI Overviews not to cite me?
Partly. A noindex directive or blocking Googlebot in robots.txt excludes you from all Google results, including AI Overviews. A less drastic option is controlling how you are shown: the nosnippet and max-snippet directives and the data-nosnippet attribute limit how much text Google may use in snippets and AI features, while the page stays in the organic results. The Google-Extended token does not control AI Overviews. Source: AI features and your website.
Is it better to be in AI Overviews or in first place on Google?
Ideally both — and they don't conflict. Pages that Google cites in AI Overviews usually rank in positions 1–5 organically. AI Overviews gives you extra visibility on top.
Why does AI Overviews cite us for some queries but not others?
AI Overviews is selective. For each query, Google evaluates whether it has enough quality sources to synthesise an answer. If not, no overview is shown. For a category where your content is weak, AI Overviews will not generate an answer that cites you.
Do review systems (Heureka, Google reviews) help with AI Overviews?
Yes, significantly. AggregateRating in schema.org is one of the strongest trust signals. An online store with 200+ reviews and a rating of 4.5+ has a much higher chance of being cited than a store with no reviews.
Conclusion
Google AI Overviews is not a trend; it is the new reality of e-commerce in Slovakia. 30–40 % of searches now start with an AI summary rather than a list of links. An online store that doesn't invest in schema.org, FAQ content and entity building will gradually lose organic visibility.
The good news: most competitors aren't dealing with this yet. Start with a free audit at pozorio.sk — you'll find out which queries you appear (or don't appear) for in AI Overviews, ChatGPT and Gemini. You'll see the first results in 60 seconds.