How to Check If Your Brand Is Cited in Google AI Overview

Ranking is not the same as being cited

I kept hitting the same problem. A page would rank on the first results page, but Google's AI Overview at the top of the search would answer the question using three other sites and never mention it. The rank report said everything was fine. The AI answer said otherwise. I wanted a number for "is my brand in the answer," so I built an API for it.

The Google AI Overview API sends any query to Google's AI Overview and returns the AI answer plus every cited source as structured JSON.

What it returns

One clean row per query, with:

• ai_overview_present: whether Google showed an overview at all.

• text_blocks: the answer as ordered paragraphs, lists, and tables.

• references: each cited source with its title, link, snippet, and domain.

• gl, hl, location: the country, language, and location used for the search.

When Google defers the overview behind a short-lived token, the API resolves the full answer inside the same run.

[[

IMG1]]

Run it in three steps

Open the Actor, enter a query or a list of brand queries, and optionally set the country and language. Run it, then read the dataset as JSON, CSV, or Excel. Pricing is pay-per-event: $0.01 to start a run plus $0.015 per retrieval.

Use it as a REST API

Start a run through the Apify API and read the dataset it produces:

POST /v2/acts/johnvc~google-ai-overview-api/runs
{"queries": ["best crm for startups"], "gl": "us", "hl": "en"}

Wire it into Claude with MCP

The Actor is MCP-ready, so Claude and other MCP clients can call it as a tool. Add its Actor-specific MCP URL, then ask your agent "is my domain cited in the Google AI Overview for these ten queries" and it reads back the answers and the sources.

[[

IMG2]]

What people build with it

Brand-citation dashboards, keyword source maps, weekly change reports, and agent workflows that flag when a domain drops out of the answer for a tracked query.

Example tasks you can run

Three runnable examples ship with the API: check if your brand appears in Google AI Overviews, see which sites Google AI Overview cites for a keyword, and monitor Google AI Overviews from Claude via MCP. Browse them on the Google AI Overview API project.

FAQ About Scraping Google AI Overviews

Where does the Google AI Overview scraper get its data?

It reads Google's own AI Overview for the query you send and returns the answer text and the sources Google cites, as structured JSON. Coverage is limited to where Google shows an overview: mainly question-style English searches in a set of countries. Queries with no overview come back flagged, so you always know.

How much does the Google AI Overview API cost per query?

Pricing is pay-per-event: $0.01 to start a run plus $0.015 per retrieval. An inline answer is one retrieval, so a single query is about $0.025; a deferred answer needs two. A batch of ten inline answers runs about $0.16. You call it over REST and read the dataset as JSON, CSV, or Excel.

Can I use the Google AI Overview scraper from Claude with MCP?

Yes. The Actor is MCP-ready, so any MCP client can call it as a tool. Add the Actor-specific MCP URL and ask your agent to check whether your domain is cited for a set of queries; it returns the answers and references with no code.

Can I schedule the Google AI Overview scraper to run automatically?

Yes. Save a task with your query set and attach a schedule from the Actor's Actions, then Schedule menu. Daily at 7 AM is the cron string 0 7 * * *. One schedule can trigger many tasks, so you can watch dozens of brand or keyword lists from a single timer and build a citation history over time.

Issues get a same-day look, so if a query returns something unexpected, send it over and it gets fixed.