Gemini, Claude and Copilot: How Each One Finds Brands

Gemini, Claude and Copilot work on the same broad pattern: a model with built-in knowledge up to a cutoff date, plus web search for anything newer or more specific. Where they differ is in who runs the search, which crawlers you can control, and how much each vendor documents. For a brand, the foundation is shared: be described accurately, in context, on pages these systems can crawl and retrieve.

Each section below states what the vendor documents, then any independent studies, then our reasoning. Where something is not documented, we say so. For ChatGPT and Perplexity, see our separate articles on where ChatGPT gets its information and how Perplexity picks its sources.

The three at a glance

Gemini: how does Google’s assistant find brands?

Documented facts. Google said in July 2026 that the Gemini app had 950 million monthly active users, and by August 2026 reports said it had passed 1 billion monthly users. Google has not published how it defines a monthly user.

Google’s crawler documentation says Google-Extended governs whether content is used to train future Gemini models and for grounding in the Gemini apps and Vertex AI. It “does not impact a site’s inclusion in Google Search” and is not a ranking signal.

On training data, the Gemini 2.5 Pro model page lists a knowledge cutoff of January 2025. Newer Gemini model pages did not list a cutoff when we checked, so we cannot say how recent their built-in knowledge is.

Reasoning. Gemini sits close to Google Search. The pages that help you in Google AI Overviews and AI Mode, indexed, snippet-eligible and widely describing your brand, are the natural starting point for Gemini too. Blocking Google-Extended trades Gemini grounding for control over training; most brands seeking visibility leave it open.

Claude: how does Anthropic’s assistant find brands?

Documented facts. Claude gained web search in March 2025, and since May 2025 web search is available globally on all Claude plans. Anthropic documents three crawlers in its help center:

• ClaudeBot collects web content that could contribute to training. Blocking it signals your future content should be excluded from training datasets.
• Claude-SearchBot indexes content for search. Disabling it “prevents our system from indexing your content for search optimization.”
• Claude-User retrieves pages when a person’s query needs them. Disabling it prevents Claude from retrieving your content in response to a user query.

Anthropic says its bots honor robots.txt and the Crawl-delay extension. Its newest models list a training data cutoff of June 2026.

Not verified. Press reports in 2025 named the search provider behind Claude’s web search, but we could not confirm it from a primary source, so we do not state it.

Reasoning. Like OpenAI, Anthropic documents separate bots for training and for search. A brand can block ClaudeBot and still be found in Claude’s search answers, as long as Claude-SearchBot and Claude-User are allowed. If your robots.txt blocks “anthropic” or “claude” broadly, check which bots that actually covers.

The June 2026 cutoff also matters. Anything about your brand published after that date, such as a launch, a new product line or a move into a new country, is not in the model’s built-in knowledge. Claude can only learn it by searching, which brings the question back to whether accurate pages about you exist and can be indexed.

Copilot: how does Microsoft’s assistant find brands?

Documented facts. A Microsoft spokesperson told TechCrunch that Copilot had about 150 million monthly active users across its commercial and consumer versions, and Satya Nadella said consumer daily users grew nearly three times year over year.

Microsoft also reported 15 million paid Microsoft 365 Copilot seats, which means a large share of Copilot use happens at work, inside Word, Outlook and Teams. Reasoning. For B2B brands this matters: a buyer drafting a vendor shortlist in a work tool may ask Copilot rather than open a browser.

We have not verified a current Microsoft document describing how Copilot retrieves web content or which crawlers govern it, so we do not describe its internals here.

Reasoning. Microsoft runs Bing, and Bing has long been a search source for AI assistants: Microsoft announced it as ChatGPT’s default search experience in 2023. It is sensible to make sure your site and the pages about you are indexed in Bing, using Bing Webmaster Tools, rather than checking Google alone. That costs little and helps across several engines.

How accurate are these assistants about what they find?

Independent research. The European Broadcasting Union and 22 public broadcasters tested ChatGPT, Copilot, Gemini and Perplexity on news questions in 14 languages. They found 45% of answers had at least one significant issue, and Gemini had the highest rate, 76%. That study was about news, not brands, but it shows that retrieval does not guarantee accuracy. A hallucination about your company is possible on any engine.

Answers also vary between runs. A SparkToro study across ChatGPT, Claude and Google’s AI found less than a 1 in 100 chance of the same brand list appearing in any two runs. Measure your mention rate across many runs, not one answer.

Do you need a different plan for each engine?

Study plus reasoning. Ahrefs found brand visibility outputs correlate at 0.749 to 0.821 between AI Overviews, AI Mode and ChatGPT. Copilot, Claude and Gemini were not in that study, but the pattern suggests one shared foundation does most of the work:

• Crawlers allowed for search on every engine you care about.
• Your site indexed in both Google and Bing.
• Consistent brand facts everywhere, including entity data like Organization schema.
• Accurate brand mentions in context on many indexed publications.

Per-engine work is then mostly measurement. Suppose a boutique hotel in Lisbon checks “quiet boutique hotels in Lisbon for couples” in each assistant five times. It may find it is named often in Gemini, rarely in Claude and never in Copilot, which points to where to look next.

What to do next

• Review robots.txt for Google-Extended, ClaudeBot, Claude-SearchBot and Claude-User, and decide on each deliberately.
• Verify your site in Bing Webmaster Tools as well as Google Search Console.
• Get a free AI visibility check to see how Gemini and the others describe you today, or read about The Blue Ocean Overview for work focused on Gemini and Google’s AI answers.
• For brands ChatGPT already mentions, The Blue Ocean GPT focuses on how consistently and how high it names them.