How to get your brand named in ChatGPT, Claude, Gemini and Perplexity answers
The answer rarely comes from your own site. Six checks decide whether an assistant names you.
Where an assistant’s answer comes from
An assistant builds its answer from pages it can fetch and trust. Most of them are not yours; your site supplies the facts they are checked against.
Embryo, on ChatGPT mentions: it’s crucial to look outside of your own website
(source).
Can the assistants read you at all?
Each assistant fetches with a named crawler, and one line in robots.txt decides whether it gets in. Our own file is the worked example.
Dageno: Blocking crawlers like GPTBot or Google-Extended reduces your chances
(source).
kanavulab.com/robots.txtUser-agent: *Allow: /Sitemap: https://kanavulab.com/sitemap.xml# AI Crawlers - explicitly allowed for GEOUser-agent: GPTBotAllow: /User-agent: ChatGPT-UserAllow: /User-agent: PerplexityBotAllow: /User-agent: ClaudeBotAllow: /User-agent: Google-ExtendedAllow: /
- GPTBot allowedOpenAI
- ChatGPT-User allowedOpenAI, on request
- ClaudeBot allowedAnthropic
- PerplexityBot allowedPerplexity
- Google-Extended allowedGemini
- Everyone else, by the * ruleOAI-SearchBot, Bingbot, newer agents; sitemap declared
Answer-first pages: one page per buyer question
One buyer question per page, the question as the heading, the answer in the first lines. Then the facts, then the FAQ.
Consistent facts everywhere
Name, what you do, where you are: the same words on your site, your Google profile, LinkedIn and every directory. A mismatch is a reason to leave you out.
Measuring it weekly
The same questions to the same assistants every week, each in a fresh session. Named means an answer names or cites you.
On our own brand: ChatGPT named KanavuLab in 17 of 33 answers to one tracked buyer question between 28 September and 10 October 2026.
One question, one assistant, counted in answers, not days.
KanavuLab, our own brand
best MVP development company in Chennai for a startup founder with an app idea
What we do about it
We do
- Track your buyers’ questions weekly, in fresh sessions.
- Write one answer page per question, facts matching your profiles.
- Open crawler access, add schema, report named out of tracked.
We never
- Promise a mention. Assistants decide.
- Sell a placement. Citations are not bought.
- Hide a miss. The number counts them.
Scope and pricing: the AI SEO agency page.
Questions
Does answer engine optimization replace SEO?
No. Both rest on crawlable pages, clear facts and trusted mentions. Answer engine optimization adds one page per buyer question and a weekly check of the answers.
Can we pay to be mentioned in ChatGPT answers?
No placement that we know of. Assistants build answers from pages they fetch and trust.
How long before a brand gets named?
No fixed time, and no guarantee. We report the weekly number from the first week, misses included.
Which crawlers should we allow in robots.txt?
GPTBot, ChatGPT-User and OAI-SearchBot (OpenAI), ClaudeBot (Anthropic), PerplexityBot, Google-Extended (Gemini), and an open general rule for Bingbot and newer agents. Our robots.txt is the example above.
Does this work for a small business in Chennai?
Where buyers ask an assistant before they buy, yes. The KanavuLab result above is a Chennai question.
What does it cost?
Scoped and quoted after a 20-minute call; scope on the AI SEO agency page.
Related
Find out if you are named today.
Scoped and quoted after a 20-minute call. Not the right fit? We will say so.