GEO audit
A GEO audit is a review of how visible a brand is in generative AI answers, and why. It checks whether crawlers such as OAI-SearchBot and Googlebot can reach the site, whether ChatGPT, Gemini, Microsoft Copilot and Google AI Mode name the brand for key prompts, and which pages those answers cite.
Why a GEO audit matters
A GEO audit often finds problems nobody on the marketing team knew about, starting with access. AI crawlers can be blocked by a setting that lives outside your robots.txt. In September 2024, Cloudflare, which says it handles traffic for 20% of the web, added a one-click option to block AI crawlers, and by July 2025 more than one million customers had turned it on. Since July 1, 2025, every new domain on Cloudflare is asked upfront whether to allow AI crawlers at all.
Google flags the same risk for its own AI features: crawling must be allowed "in robots.txt, and by any CDN or hosting infrastructure". A brand can publish excellent pages and still be missing from some engines because of a single firewall rule.
What a GEO audit checks
A complete GEO audit follows the path from your server to the answer:
- Crawler access. Robots.txt rules and CDN settings for OAI-SearchBot (ChatGPT search), Googlebot (Google AI Mode) and Bingbot (Microsoft Copilot). OpenAI's GPTBot only collects training data, so blocking it does not remove you from ChatGPT search.
- Readability. Whether key content is in the HTML. Vercel found that GPTBot, ClaudeBot and PerplexityBot don't execute JavaScript, unlike Googlebot. Technical GEO covers this in detail.
- Presence. Whether each engine names the brand for the prompts buyers ask, how often and in which position.
- Competitors. Which brands are named instead.
- Sources. Which pages the answers cite, and whether they mention the brand.
- Accuracy. Whether prices, features and positioning are described correctly.
One-off audit vs continuous tracking
A one-off GEO audit makes a good baseline, especially for technical issues: a blocked crawler stays blocked until someone fixes it. The answers move much faster. In a January 2026 SparkToro study, City of Hope appeared in 69 of 71 ChatGPT answers about the best cancer care hospitals on the U.S. West Coast, yet it was the top mention in only 25 of them. A single screenshot could show it first or far down the list.
The two formats answer different questions. An audit tells you what to fix now. Tracking the same prompts over weeks tells you whether the fixes worked and when competitors move. Most brands need the first to start and the second to keep up.
How to run a GEO audit
Start with access: test your robots.txt against each AI user agent, review the bot settings of your CDN or firewall, and load key pages with JavaScript turned off. Then choose a few dozen prompts your buyers actually ask, run each one several times on every engine, and record mentions, positions and cited pages for you and two or three competitors.
Hawkscan's action plan works like a GEO audit that keeps running. It reads the answers from ChatGPT, Microsoft Copilot, Gemini and Google AI Mode every day, along with their sources and your site, and lists fixes ranked by impact, such as an AI crawler your site blocks or a page that cites competitors but not you. Your AI visibility score shows whether they pay off.
Sources
- Cloudflare, Cloudflare Just Changed How AI Crawlers Scrape the Internet-at-Large; Permission-Based Approach Makes Way for A New Business Model ()
- Google Search Central, AI features and your website ()
- OpenAI, Overview of OpenAI Crawlers
- SparkToro, NEW Research: AIs are highly inconsistent when recommending brands or products; marketers should take care when tracking AI visibility ()
- Vercel, The rise of the AI crawler ()