How to show up in ChatGPT, Claude, Gemini and Perplexity answers

To show up in ChatGPT, Claude, Gemini and Perplexity answers you need three things: their systems must be able to reach your pages, they must understand exactly who you are and what you offer, and they must find independent confirmation of your value outside your own site. Each assistant has its own sources and logic, so generic "optimise for AI" advice produces random results. This guide explains how each engine works and the actions that pay off across all of them.

Two ways an AI talks about you

• Model knowledge: what the model learned during training. If your brand was mentioned often and consistently in public text (articles, forums, reviews, Wikipedia), the model may name you without searching the web. This changes slowly, with new model releases.
• Live retrieval: the assistant runs a web search, reads a few pages and summarises them with citations. This is how ChatGPT search, Perplexity, Copilot, Gemini and Google AI Overviews work. Indexing, rankings and passage quality matter here, and improvements show up within weeks.

An effective strategy works on both: content and technical signals for retrieval, reputation and mentions for model knowledge.

How each assistant searches

ChatGPT

When search is enabled, ChatGPT relies on third-party search indexes (Bing has historically been a major one) and on OpenAI's own OAI-SearchBot crawler. OpenAI uses GPTBot for training data and ChatGPT-User when a user asks it to open a page. As OpenAI's crawler documentation explains, you can block training without giving up search visibility. Ranking well on Bing is a direct advantage: verify your site in Bing Webmaster Tools and submit your sitemap.

Claude

Anthropic's Claude can search the web and uses its own crawlers, such as ClaudeBot and dedicated search user agents. It favours sources with clear, verifiable information and often cites official documentation and reference pages. Well-structured product pages and guides with up-to-date facts work best.

Gemini and AI Overviews

Both are built on Google's index, so classic SEO rules apply: correct indexing, helpful content, E-E-A-T signals, structured data. The Google-Extended token in robots.txt lets you opt out of Gemini training without affecting Google Search rankings. Pages that rank well for related queries have a good chance of being cited in AI Overviews.

Perplexity

Perplexity is the most transparent engine: it always shows numbered sources and usually cites many of them. It uses its own PerplexityBot crawler and favours fresh content, direct answers and niche industry sources. For many B2B companies it is the first assistant where results become visible.

8 actions that work on every engine

1. Let the right crawlers in. Check robots.txt, firewall rules and CDN bot protection, then confirm in your server logs that the bots actually get through – "bot fight" features often block AI agents even when robots.txt allows them. With the SEOGenio log analyzer you can upload server logs and see which AI crawlers visit, which pages they read and how often.
2. Make content readable without JavaScript. Many AI crawlers do not execute JavaScript. Prices, product descriptions, FAQs and main copy must be in the server-rendered HTML.
3. Lead with a direct answer. Open every important page with 2–3 self-contained sentences that answer the main question – the format engines extract most readily.
4. Structure by sub-question. Use H2/H3 headings phrased like real user questions. Each section should stand on its own, because it may be quoted in isolation.
5. Add data, examples and sources. Vague claims ("the best solution on the market") are ignored; verifiable ones ("daily tracking of 500 keywords with 16 months of Search Console history") get quoted.
6. Use structured data. Organization with logo and sameAs profiles, Product or SoftwareApplication with prices, Article with author and date, FAQPage where it makes sense.
7. Earn third-party mentions. Assistants build "best tools" lists from comparison articles, reviews and discussions. Find the pages cited for your key prompts and work to be included.
8. Publish an llms.txt file. A Markdown summary of your site and its key links at the domain root. It is an emerging proposal and not every engine reads it, but it is cheap and helpful for AI agents – see our llms.txt guide.

How to check that you are improving

Checks must be systematic, or you end up judging from a couple of lucky conversations:

1. Pick a fixed set of at least 20 prompts split by intent: informational, comparative, transactional.
2. Run the same prompts on every model at regular intervals, ideally weekly.
3. For each answer record mention yes/no, position in the list, sentiment and cited URLs.
4. Compare against 3–5 direct competitors: visibility is always relative.

The SEOGenio AI Visibility module runs these checks automatically on ChatGPT, Claude, Gemini and Perplexity, calculates your share of mentions versus competitors and shows which sources are cited instead of you, so you know exactly where to act.

How long does it take?

For retrieval-based engines (Perplexity, ChatGPT search, AI Overviews) the first citations can arrive within a few weeks, as soon as updated pages are recrawled and rank. Model knowledge only changes with new releases, so reputation work pays off in the medium term (6–12 months). That is why it pays to start on both levers now.

New to the topic? Read our introduction to what GEO is and why it matters as much as SEO, then start your free trial and run your first visibility check on your own prompts.