AI answer engines recommend companies the same way they answer any other question: they run searches, read the pages that come back, and write an answer from what those pages say. A company gets named when it appears clearly and consistently in the pages the engine retrieves for that question, whether those are its own site or someone else's. There is no submission form and no separate ranking to buy; there is retrieval, and there is what the retrieved text says about you.
How does an answer engine find companies to name?
By searching, several times, and reading the results. Google states that AI Mode and AI Overviews may use a "query fan-out" technique, issuing multiple related searches across subtopics and data sources, to develop a response. A question like "best invoicing software for a small law firm" becomes a handful of searches, and the pages that rank for those searches are the raw material of the answer.
The other engines document the same shape through their crawlers. OpenAI's OAI-SearchBot is used to surface websites in search results in ChatGPT's search features, and when a user asks a question, ChatGPT-User may visit a web page to answer it. Perplexity's PerplexityBot is designed to surface and link websites in search results on Perplexity, and its Perplexity-User fetcher might visit a web page to help provide an accurate answer. Anthropic's Claude-SearchBot indexes content for search responses and Claude-User accesses websites when individuals ask Claude questions.
The consequence is simple. If the engine cannot retrieve a page that names you, you do not exist for that question, no matter how good the product is.
What does Google require to appear in AI Overviews and AI Mode?
Only what it already requires for search. Google's documentation says that to be eligible as a supporting link, a page must be indexed and eligible to be shown in Google Search with a snippet, and that there are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary. It goes further: you don't need to create new machine readable files, AI text files, or markup, and there is no special schema.org structured data you need to add.
What it does recommend is ordinary hygiene: allow crawling in robots.txt, make important content available as text, and make sure structured data matches the visible text on the page. It also confirms that the usual preview controls apply, so nosnippet, data-nosnippet, max-snippet and noindex limit what an AI feature can show from your page.
Which crawlers do you need to allow?
Each engine separates the crawler that builds its search index from the one that collects training data, and blocking the wrong one removes you from answers without protecting anything.
| Engine | Allow for answers | User-triggered fetcher | Training crawler (separate) |
|---|---|---|---|
| ChatGPT | OAI-SearchBot | ChatGPT-User | GPTBot |
| Perplexity | PerplexityBot | Perplexity-User | none stated; PerplexityBot is not used to crawl content for AI foundation models |
| Claude | Claude-SearchBot | Claude-User | ClaudeBot |
| Google AI Overviews and AI Mode | standard Search crawling, allowed in robots.txt | not separately documented | not covered on the AI features page |
OpenAI is explicit about the cost of a mistake: sites that are opted out of OAI-SearchBot will not be shown in ChatGPT search answers, while disallowing GPTBot only indicates a site's content should not be used in training. Two of the user fetchers also do not honor robots.txt, since a person asked for the page: OpenAI says robots.txt rules may not apply to ChatGPT-User, and Perplexity says its Perplexity-User fetcher generally ignores robots.txt rules.
What makes a retrieved page get cited?
Being quotable. The paper that coined generative engine optimization, released in November 2023 and accepted at KDD 2024, tested nine content changes on a benchmark of 10,000 queries and found that adding quotations produced a relative improvement of 30 to 40% on its position-adjusted word count metric, with statistics and cited sources close behind, while keyword stuffing offered little to no improvement. The abstract's headline is that GEO can boost visibility by up to 40% in generative engine responses.
Read the caveat with it. A July 2026 survey of GEO research concludes that topical relevance and context position are the most reproducible levers, that no reviewed technique shows a stable, longitudinal, cross-platform causal effect on organic discoverability, and that citation-oriented rewrites can impair retrieval. In other words: a page must be retrieved before any of the citation tricks matter, and a page rewritten purely to be quotable can rank worse in the search that feeds the engine.
The practical order is therefore fixed. First, answer the question the query asks, plainly, near the top of the page. Then make the answer easy to lift: a specific number, a named source, a sentence that stands on its own.
Where do the citations actually go?
Mostly to companies' own sites, and then to comparison pages. A June 2026 study by Ranqo, a visibility tracking vendor, analyzed more than 100,000 prompt responses across more than 100 brands between March and May 2026 and reported that about 78% of citations go to corporate websites, with YouTube leading the non-corporate sources ahead of Reddit, editorial media and Wikipedia. The most cited content format was "best of" listicles at approximately 21% of all citations.
The same study describes a three-tier hierarchy: global household names appeared in 73% of relevant AI answers, established mid-market and regional brands in 44%, and niche and small brands in 11%. It also found that sentiment flips about 6.7 times more often than whether a brand is mentioned at all. This is one vendor's data, and it should be weighed as such, but its two lessons match what the engines' own documentation implies: your site is the largest citation pool you control, and the listicles and comparison articles that rank for "best X" queries are the second.
That second pool is where earned coverage does real work. A placement on a publication that already ranks for your category's comparison queries is a page the engine retrieves, and if it names you with a specific claim, it becomes a citation. That is the logic behind the placements catalog at Qubba Publishings & PR, and behind the content work at Qubba Marketing, which writes the pages on your own domain to be retrieved and quoted.
What should a company do this quarter?
Six things, in order. Allow OAI-SearchBot, PerplexityBot and Claude-SearchBot in robots.txt, and confirm your pages are indexed with snippets in Google. Publish one page per question you want to be the answer to, with the answer in the first paragraph. Put a real number and a named source in each of those pages. Check that your structured data says what the visible page says. Get named, accurately, on the comparison pages that already rank for your category. Then ask each engine your ten most valuable questions once a month and write down who it names and how.
Frequently asked questions
Does blocking GPTBot remove my company from ChatGPT answers?
No.
OpenAI documents GPTBot as the training crawler and OAI-SearchBot as the search crawler, and says disallowing GPTBot indicates a site's content should not be used in training while opting out of OAI-SearchBot removes a site from ChatGPT search answers.
You can block training and stay visible in answers.
Do I need schema markup or an llms.txt file to appear in Google AI Overviews?
Not according to Google, which states that you don't need to create new machine readable files, AI text files, or markup and that no special structured data is required.
Structured data still helps ordinary search, and Google asks that it match the visible text on the page.
Why does a competitor get recommended when my product is better?
Because the engine retrieved pages that describe the competitor and did not retrieve pages that describe you, or retrieved pages that describe you vaguely.
Research on generated answers found that quotations, statistics and cited sources lifted visibility while relevance and position decide what gets retrieved at all.
Fix retrieval first, then specificity.
Can a company pay to be recommended by an AI answer engine?
None of the crawler or eligibility documentation from Google, OpenAI, Perplexity or Anthropic linked above describes a paid placement inside the answer itself.
What you can do is pay for coverage on publications the engines already retrieve, and make sure that coverage states specific, checkable facts about you.