AI Overviews
AI Overviews are the AI-written summaries Google shows at the top of some search results, with links to the pages it used. Search "how to calculate CAC payback" and you'll often see one above the regular results, so being one of the linked sources matters more than ranking just below it.
What are AI Overviews?
AI Overviews are the short AI-written summaries Google shows at the top of some search results pages. Each one answers the search in a few sentences or a short list, with link cards to the pages it used. The regular results sit below it.
Let's say a finance lead at a SaaS startup searches "how to calculate CAC payback". Before any blue link, Google shows a summary: the formula, a quick example and a note on what counts as a good number. On the right sit links to three or four pages it drew from. If your guide is one of them, the finance lead sees your brand before she reads a single result. If it's not, your number one ranking now sits below a box that already answered her.
Google launched AI Overviews in the US in May 2024, after testing them as the Search Generative Experience in 2023. In May 2025 it took them to more than 200 countries and territories and over 40 languages. They're different from AI Mode, which is a separate tab for a full chat-style answer.
Why AI Overviews matter for B2B SaaS
AI Overviews reach more people than any other AI feature. Google says 2.5 billion people use them each month. And they cut clicks: in a Pew Research Center study, people clicked a regular result on 8% of searches with an AI summary, against 15% without one.
That's a big deal for SaaS teams that built their pipeline on "what is" and "how to" content. Those are the searches where AI Overviews show up most. Here's what changes:
- Top rankings lose clicks: A page can hold position one and still lose a lot of traffic when an AI Overview sits above it. Ranking reports alone will tell you everything is fine.
- The citation becomes the prize: The pages linked inside the overview get the attention that used to go to the first result. Being one of three cards beats ranking fourth below the box.
- Your brand gets described for you: Branded searches often show an overview too. What it says about your pricing and product is what a buyer reads first.
How to get cited in AI Overviews
Google says a page only needs to be indexed and eligible for a snippet to appear. There's no special markup. So the work is mostly good SEO aimed at the questions behind each overview. Here's my process:
- Find the searches that show one: Check each target search in a private window, then try longer question versions. In Pew's study, only 8% of one or two word searches got an AI summary, against 53% of searches with 10 or more words.
- Rule out technical blocks first: Check for noindex, nosnippet and low max-snippet values in your page templates. Then use URL Inspection in Search Console to confirm Google sees your main text after rendering.
- Answer under each heading: Write headings as the questions buyers search and answer each one in the first sentence or two. Google pulls short, complete answers far more easily than ones spread across paragraphs.
- Cover the related questions: Google builds each overview from several related searches. A page that also answers "what's a good CAC payback for SaaS?" and "CAC payback vs LTV to CAC" gets more chances to be pulled in.
- Get named on other sites: Ahrefs found branded web mentions had a 0.664 correlation with AI Overview mentions across 75,000 brands, against 0.218 for backlinks. Reviews, roundups and guest articles all count.
Our post on why brands miss AI Overviews walks through ten causes in the order I'd check them.
AI Overviews examples
The best way to learn how AI Overviews pick sources is to look at the types of searches B2B buyers run. Here's how they tend to differ:
| Search type | Example | What the overview usually does | What to build |
|---|---|---|---|
| Definition | "what is SOC 2 type 2" | Defines the term and lists the main parts | A definition page with a one-sentence answer up top |
| How-to | "how to calculate CAC payback" | Gives the formula and steps | A guide with the formula, a worked example and a calculator |
| Comparison | "jira vs linear" | Compares features and who each fits | A fair comparison page with a table and a "last checked" date |
| Branded | "[your brand] pricing" | Summarizes plans from your site and others | A pricing page with plan names and prices in HTML text |
Semrush found AI Overviews grew 71% on commercial results between November 2025 and April 2026, so comparison searches like the third row are more exposed every quarter.
How to measure AI Overviews
Search Console doesn't split out AI Overview data, so you have to piece it together. Here's what I'd track:
- Overview presence by keyword: Use a rank tracker that flags which of your target searches show an AI Overview and whether your page is cited in it. Check monthly, since the share of searches with one keeps moving.
- Click-through rate on affected pages: In Search Console, compare CTR before and after an overview appeared for each page's main queries. Falling CTR with steady impressions is the classic sign.
- Citations by page: Keep a list of which of your pages are cited in which overviews. That tells you which page formats Google likes to quote.
- Branded overview accuracy: Search your own brand, your pricing and "[your brand] vs [competitor]" every month. Fix the page behind any wrong fact.
Common AI Overview mistakes
- Reading rankings as success: A page can rank first and lose most of its clicks under an overview. Track CTR and citations next to position.
- Adding schema to get in: Google doesn't need special markup for AI Overviews. Spend that time on clearer answers and fixing technical blocks.
- Turning on nosnippet in a panic: It does keep you out of AI Overviews, but it also strips your snippet from the regular results. For most B2B pages that's a bad trade.
- Writing long intros: Three paragraphs of background before the answer make a page hard for Google to pull from. Put the answer first.
- Changing dates without changing facts: Google's helpful content guide warns against freshening dates when nothing changed. Update the facts, then the date.
Here's a first step for this week. Search the five "how to" and "what is" queries that bring you the most traffic and see which ones show an AI Overview. Are you one of the linked sources?
Read more
FAQ
How do I get my page into Google AI Overviews?
Make sure the page is indexed and allowed to show a snippet, then answer the question in the first sentence under a clear heading. Ranking for the main search and its related questions helps, and so do mentions of your brand on other sites.
Can I opt out of AI Overviews?
Only by limiting snippets. Google says the nosnippet rule keeps content out of AI Overviews and AI Mode, but it also removes the text snippet from your regular listing, so most B2B sites shouldn't use it.
Do AI Overviews show up for every search?
No. They show most often on longer questions and less on short keywords. Google also changes how often they appear from month to month, so check your own target searches regularly.
Do I need schema markup for AI Overviews?
No. Google lists no special markup or AI text files for its AI features. Structured data that matches your visible text can still help Google understand the page.



