Aleksey Klempner
TL;DR
- An AI visibility audit checks whether AI assistants can reach, read and quote your site. That part is technical and fixable.
- It cannot tell you whether those assistants will recommend you. You find that out by asking them, on a schedule.
- In our own October test, AI engines named CTO.LA in 21 of 24 answers about us or "CTO in Los Angeles," and in 0 of 24 answers to broader category questions.
The trend: the answer comes before the click
Search used to hand people ten links. Now it often hands them a paragraph. In a Pew Research Center analysis (July 2025) of 900 U.S. adults' browsing in March 2025, 18% of Google searches produced an AI summary. When a summary appeared, people clicked a traditional result in 8% of visits, against 15% when no summary appeared. They clicked a link inside the summary in just 1% of visits.
Pew also found that longer, question-style searches trigger summaries far more often: 60% of queries starting with "who," "what," "when" or "why" got one. Those are the queries buyers type when they are choosing a vendor. So the question for a business is no longer only "do we rank?" It is "when the assistant answers, are we in the answer?"
That is the gap an AI visibility audit is supposed to close. Most people searching for one ("ai visibility audit free," "how to do ai visibility audit," per Google autocomplete) want a quick score. A score is useful. It is also only half the picture.
How it works
An audit and a citation check measure two different things. The first asks "can AI systems get to us?" The second asks "do they actually name us?"
What an audit can tell you
The technical half is well documented, and you do not need guesswork for it.
- Access. OpenAI's crawler documentation says sites that opt out of OAI-SearchBot "will not be shown in ChatGPT search answers." GPTBot, the training crawler, is a separate setting. Plenty of sites block one while meaning to block the other, or a CDN bot rule blocks both without anyone deciding to.
- Eligibility in Google. Google's AI features guide says a page must be indexed and eligible to show with a snippet to appear as a supporting link in AI Overviews or AI Mode, and that "there are no additional technical requirements."
- Readable text. The same guide lists "making sure that important content is available in textual form" and "making sure your structured data matches the visible text on the page." A site that renders everything with JavaScript can look fine in a browser and nearly empty to a crawler.
Our free AI Visibility Score checks this kind of readiness. It is a good first step.
What an audit can't tell you
An audit cannot tell you whether an assistant will pick you over a competitor. Google says plainly that meeting every requirement "doesn't mean that Google will crawl, index, or serve its content." Being eligible is not the same as being chosen.
Here is our own data. On October 16, 2026 we ran 12 real buyer questions through four engines: ChatGPT, Gemini, Perplexity and Google AI Overviews. That makes 48 answers, each with a screenshot.
| Question type | Answers naming CTO.LA |
|---|---|
| About us by name, plus "CTO Los Angeles" style questions (6 questions) | 21 of 24 |
| Broader category questions: "AI website audit," "free AI visibility audit for my website," "what is agentic commerce," "fractional CTO for e-commerce / Shopify brands" and two more (6 questions) | 0 of 24 |
The site passes the technical checks, and the engines clearly know it exists. For "free AI visibility audit for my website," which is literally a product we offer, every engine named someone else. One engine also got a basic fact about us wrong. No crawler setting fixes either of those.
That gap is usually about content and outside sources: does a page answer the exact question in the first lines? Is the business mentioned anywhere besides its own site? Pew found Wikipedia, YouTube and Reddit were the most-cited sources in Google's AI summaries, together 15% of the sources listed. Assistants lean on places other than your homepage.
A caveat on method: we ran one snapshot from one location (Google placed us in Virginia, not LA), and one engine was signed in. A single run shows where you are. It doesn't show a trend. That's why we repeat it.
What a CTO should do this quarter
- Check access in five minutes. Open your robots.txt and your CDN's bot settings. Decide on purpose whether OAI-SearchBot, GPTBot and Googlebot are allowed. Search and training are separate choices.
- View your key pages as a crawler does. Fetch them without JavaScript (for example with
curl). If the main copy or service names are missing, fix rendering before anything else. - Make structured data match the page. Organization, service and FAQ markup should repeat what visitors can read, not add claims they can't see.
- Write down 10–12 real buyer questions. Use your sales calls and autocomplete, not your brand name. Include the questions where you should win but aren't known yet.
- Run them through four engines, monthly. Log answer shown, named or not, cited URL and competitors named, with a screenshot. Same questions, same method, every time.
- Fix the gaps that the panel shows. For each question you lose, publish one page that answers it directly in the first lines. Then earn one mention somewhere outside your site.
- Measure with what you already have. Google says AI feature traffic is counted in Search Console's Performance report under "Web." Pair it with your analytics. Don't buy a dashboard before you have a baseline.
The bottom line
A free audit tells you whether the door is open. A citation panel tells you whether anyone walks through it and says your name. You need both, and the second is where most of the work is.
If you want help building that panel and closing the gaps for your own buyer questions, Book a call. Or start with the free AI Visibility Score.
FAQ
What is an AI visibility audit?
It is a check of whether AI assistants such as ChatGPT, Gemini, Perplexity and Google AI Overviews can reach, read and understand your website. It looks at crawler access in robots.txt and CDN rules, whether key content is in the HTML as text, and whether structured data matches the page. It measures readiness. It does not measure whether those assistants recommend you.
Is a free AI visibility audit enough?
It is a good start, not the whole job. An audit finds technical blockers you can fix quickly. It can't show whether you are named in real answers. In our own test, a technically sound site was named in 0 of 24 answers to category questions. You need a recurring citation check alongside the audit.
How do I check if ChatGPT recommends my business?
Write 10–12 questions your buyers really ask, without your brand name. Ask them in a fresh chat with web search on, and do the same in Gemini, Perplexity and Google. Record whether you're named, which URL is cited and which competitors appear, with screenshots. Repeat monthly with the same method.
Does blocking GPTBot hide me from ChatGPT search?
Not by itself. OpenAI says OAI-SearchBot controls whether a site appears in ChatGPT search answers, and GPTBot controls training use. They are independent settings. If you block OAI-SearchBot, OpenAI says your site will not be shown in ChatGPT search answers, though it can still appear as a navigational link.
Do I need an llms.txt file or special schema for Google AI Overviews?
Google says no. Its guide states you don't need new machine-readable files, AI text files or special schema.org markup to appear in AI Overviews or AI Mode. Standard SEO basics apply: allow crawling, keep content in text, use internal links and make structured data match the visible page.
Sources
- Pew Research Center, "Google users are less likely to click on links when an AI summary appears in the results," July 22, 2025. https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/
- Google Search Central, "AI features and your website" (accessed Oct 22, 2026). https://developers.google.com/search/docs/appearance/ai-features
- OpenAI, "Overview of OpenAI Crawlers" (accessed Oct 22, 2026). https://platform.openai.com/docs/bots
- CTO.LA, AI citation baseline, 48 answers across 4 engines, captured Oct 16, 2026 (internal data, screenshots on file).