How do I get on Google AI Overviews?

What Google AI Overviews actually does

AI Overviews is a retrieval-augmented generation layer sitting on top of Google's existing search index. It picks a small set of pages (usually 3 to 8) from the top-ranking organic results, then asks a Gemini-family model to synthesize a paragraph or two of answer. The cited URLs are linked in chips at the top of the answer block.

Two important consequences flow from this. First, you cannot be cited by AIO if you don't already rank top 10 for the query. The retrieval layer doesn't go searching the open web. It picks from what Google already ranks. Second, AIO favors pages it can ground confidently. Ahrefs' 2024 AIO study analyzed thousands of AIO answers and found citation pages were 25.7% fresher on average than the regular top 10, and disproportionately likely to have structured data and clear answer blocks at the top.

The 3 layers of optimization (rank → schema → freshness)

The order matters. Skip a layer and the next one doesn't move the needle.

Layer 1: rank. If you're position 47 for "best plumber Austin," AIO does not care about your schema. Get to top 10 first. For local-service queries, that usually means winning the standard SEO factors (backlinks, on-page relevance), plus the local-specific stuff like NAP consistency and category match in GBP.

Layer 2: schema. Once you rank, structured data is the lift that moves you from "ranks for the query" to "gets cited in the AIO answer." For a local-service page, the schema cocktail is: LocalBusiness (with full address, hours, aggregateRating, and a specific subtype like Plumber or Dentist) plus FAQPage on any page with a question-and-answer section. The Q&A structure mirrors how AIO formats its answers, which makes your content easier to lift.

Layer 3: freshness. Ahrefs found AIO citations are roughly a quarter fresher than the regular SERP. That doesn't mean rewrite every page weekly. It means update the date stamp when you do real updates, and run a quarterly refresh on your highest-value pages: pricing tables, service area lists, FAQs.

What we're seeing for local business queries

For "best [trade] in [city]" queries (the bread-and-butter local searches) AIO behaves differently from informational queries. Instead of a long synthesized answer, you usually get a one-line summary, a named business or two, and a link out to the Maps panel.

The named business is the prize. We've audited about 60 cases where a client wanted to know why a competitor was named instead of them. In every case, one of three things was true about the competitor: more reviews than the client on Google (sometimes 4-5x more), structured data that clearly identified them as the specific subtype Google expected (e.g. Plumber not just LocalBusiness), or a higher-authority cited page on their site (usually a city-specific service page with the query in the H1).

The big mistakes (long intros, no schema, stale homepages)

The pattern I see most often when an owner asks why they're invisible to AIO:

1. Long intros. The first paragraph is "Welcome to Acme Plumbing, your trusted partner for all your plumbing needs in Phoenix and the surrounding area..." AIO has nothing to lift. Replace it with a direct answer to a likely query.
2. No schema at all. WordPress sites with no SEO plugin, or a plugin installed but never configured. Run a Rich Results Test on your homepage. If it shows no eligible types, you've got work to do.
3. Stale homepage. Last-modified date from 2022. Even if nothing has changed structurally, the date stamp matters. Run a real refresh: update a stat, add a recent project, then resave.
4. Generic homepage instead of service pages. "Plumber in Phoenix" rarely ranks the homepage. Build city or service-specific pages with the exact query in the H1.

Run our free check and we'll tell you which AIO layer is your bottleneck right now: rank, schema, or freshness. Most owners are stuck on one of the three.