Page count explains 73% of how much AI crawlers visit you, and correlates 0.17 with whether you actually get named in an answer. We measure the second one. Across 9.8 million brand mentions we can show you which engines name you, which name your competitors, and what changes that.
| Crawler | What its visit tells you |
|---|---|
| GPTBot, ClaudeBot | Training data collection. Says nothing about citation. |
| OAI-SearchBot | Whether you are eligible to appear at all. |
| ChatGPT-User | A real person's question just pulled your page. |
Publishers already treat these differently. Among the top 10,000 sites, 17.6% block GPTBot but only 6.6% block OAI-SearchBot.
An AI assistant does not rank you. It retrieves a handful of sources, synthesizes an answer, and names a few companies. Each of those steps can be worked on, and they are worked on differently.
Can the engine fetch your pages when someone asks a question.
Can a model lift a clean, quotable answer out of the page.
Does the wider web name you when the engine looks for corroboration.
Per engine, every week, never blended into one score.
Each of these can make every other number on your report meaningless. We do them first, in this order, on every account.
We read your robots.txt the way each engine reads it and record a verdict per agent. A site that never named an AI crawler is a different situation from one that deliberately allowed it, and we report those differently.
OpenAI states that a site opted out of OAI-SearchBot will not appear in ChatGPT search answers. If that line is wrong, nothing else matters until it is fixed.
ChatGPT-User and Claude-User fire when a real person asks something. We separate those from bulk training crawl rather than adding them into one AI traffic number.
We record who is named on your buying questions today, per engine, so that every later claim of improvement has something honest to be measured against.
These are two different retrieval systems that happen to read the same web. Treating them as one channel is how companies end up leading on Google and absent from every assistant their buyers use.
| Question | Traditional SEO | AI search |
|---|---|---|
| What the user gets | A ranked list of links | One synthesized answer naming a few companies |
| How many results matter | Top ten, realistically top three | 3.3 brands per answer on Google AI Overviews, 6.1 to 7.2 on the chat engines |
| Overlap between the two | About 8% of ChatGPT citations rank in Google's top ten for the same prompt | |
| What moves it most | Links and on page relevance | Branded mentions at 0.664, well ahead of backlinks at 0.218 |
| How results are checked | One index, one ranking | Four engines that agree with each other 1% to 11% of the time |
Overlap figures are from Rankxa production data, 30 days to 11 September 2026. Citation and correlation figures are from published third party studies by Ahrefs and Orbit Media.
This is not run by a black box of tools. It is a team of Growth Architects who plan, build and optimize it, using AI where it accelerates the work and human judgment where it actually matters. Different specialists, one accountable team.
Meet the team →
Michael · Engineering Lead
Amal · Automation Architect
Bryn · Solutions Engineer
If we can't increase your revenue by the 6 month mark, our team works for free until we do.
“BusySeed created and managed our social media and wrote blogs that raised our SEO to a new level. We are getting new clients from organic search, and their lead generation campaigns made it easy to justify the investment by focusing on ROI.”
“BusySeed demonstrated tremendous ability in understanding our needs and taking us by the hand in the process of designing a digital strategy from scratch.”
“Great professionals. They manage our company's social media and designed our website. I highly recommend them for social media management and web design.”
AI SEO is the work of getting your company named inside the answers AI assistants give, rather than ranked in a list of links. It has three parts: making sure the engines can retrieve your pages at answer time, writing pages a model can lift a clean answer out of, and earning mentions of your company by name on sites the engines actually pull from. It is measured per engine, because the engines disagree with each other far more than most people expect.
They describe the same work. GEO, generative engine optimization and AI SEO are all names for optimizing to be named inside a synthesized answer. We use whichever term a client already uses. What matters is not the label but whether the reporting separates the engines, because a single blended visibility score hides the thing you need to act on.
Far less than you would hope. Ahrefs found only about 8% of ChatGPT citations rank in Google's top ten for the same prompt. Strong traditional SEO helps, because Googlebot activity is one of the better predictors of AI crawl volume, but it is not sufficient and it is not the same work.
It will make you more crawled, which is not the same thing. Across 2,899 domains of crawler data, site size is the strongest single predictor of GPTBot volume at 0.54, and Orbit Media found page count alone explains 73% of the variation in AI crawl attention. But in Ahrefs' study of 75,000 brands, total page count correlates just 0.17 with actual AI visibility, while branded web mentions correlate 0.664. The metric that is easy to buy is not the metric that works.
Your robots.txt, and specifically whether OAI-SearchBot is allowed. OpenAI states that a site opted out of OAI-SearchBot will not appear in ChatGPT search answers. That is binary and it sits upstream of every other thing we could measure. It is usually accidental, it takes minutes to fix, and until it is fixed the rest of the report is meaningless.
Rankxa reruns your prompt set every week across ChatGPT, Claude, Gemini and Google AI Overviews and records who gets named, in what order, on each engine separately. We track 9.8 million brand mentions across 1.29 million companies, so your competitor set is built from who actually appears on your questions rather than from who you think your competitors are.
The audit starts with the retrievability check, because it is the one finding that can make everything else on the report irrelevant. You get a verdict per crawler, a four engine baseline, and the list of companies being named instead of you.
