AI engines · Claude
Claude: how it searches, cites and recommends brands
Claude, Anthropic’s assistant, decides when to search the web and, when it does, always cites the pages it uses. To show up, your site must be reachable by Claude-SearchBot and Claude-User and appear in the sources it finds. GeoRush queries Claude directly through Anthropic’s API and its web search, localized to your country, and measures how often it names you.
Updated on

How Claude builds its answers
Anthropic’s web search tool documentation describes three steps: Claude decides when to search based on the prompt, the API runs the searches and returns results (this can repeat several times within one answer), and Claude gives a final response with cited sources. According to that page, Claude tends to search when a question depends on current or changing information, such as prices or facts about specific organizations and products, and answers directly when stable knowledge is enough. Citations are always enabled for web search and include the URL, title and a snippet of the cited text.
On crawlers, Anthropic’s help center lists three: ClaudeBot collects content that may be used for model training; Claude-User fetches pages when a Claude user asks a question that needs them; Claude-SearchBot analyzes content to improve the relevance of Claude’s search results. Anthropic says blocking the last two may reduce your site’s visibility in user searches.
Claude in GeoRush: fact sheet
| Fact | Claude |
|---|---|
| Company | Anthropic |
| How GeoRush queries it | Anthropic’s direct API with a Claude Sonnet model and the official web search tool |
| Web search | Anthropic’s own: one search per answer, with approximate location set to your brand’s country |
| Credits per answer | 6 (Claude analysis included). It is the most expensive engine to query |
| Cites sources? | Yes. GeoRush stores cited sources, the results it consulted and the search it ran |
| Key crawlers | Claude-SearchBot (search) and Claude-User (user-triggered visits) |
GeoRush configuration as of November 2026. The exact model that answered is stored with each answer.
What it means for your brand
For GEO (generative engine optimization) in Claude, this means:
- Buying questions tend to trigger search. Prices, comparisons and facts about specific products are the kind of question where, according to Anthropic, Claude searches. That is when the pages it finds count.
- Every claim comes with a source. Claude cites what it uses. If you want a claim about you to be right, the clearest, most current source has to state it.
- Search bots are not training bots. Blocking ClaudeBot affects training; blocking Claude-SearchBot or Claude-User may reduce your search visibility, according to Anthropic.
- Country matters. Searches can be localized by approximate location, so the sources Claude sees in the UK need not match those it sees in the US.
- Answers vary. Repeating each question is the only way to know how often it names you.
How to become more visible in Claude
- 1
Allow Claude-SearchBot and Claude-User
Check robots.txt and firewall rules. The guide robots.txt for AI bots has rules ready to copy.
- 2
Publish verifiable facts
Prices, plans, markets, integrations: pages with concrete, dated facts are easy to cite. See content AI cites.
- 3
Fix what’s wrong off your site
If Claude repeats a wrong fact, find the source in GeoRush and get it corrected. The guide AI wrong information about your brand explains how.
- 4
Earn mentions on third-party sources
Use Sources and citations and Get cited to appear on the sites Claude already consults.
- 5
Measure by market
If you sell in several countries, track the same question in each with Markets and languages.
How GeoRush measures your Claude visibility
- It asks Claude the questions you track, in your brand’s language and with search localized to its country, instructing it to name specific products when relevant and to cite its sources.
- It repeats each question several times (3 by default, up to 5). If a model declines a question, Anthropic’s API can hand it to another Claude model so the answer isn’t lost.
- A separate Claude analysis reads every answer: brands and order, tone, claims about you checked against your official facts, and newly recommended brands.
- It calculates Claude on its own: mention rate (with a 95% confidence range), share of voice, citation rate, average position and sentiment.
- It stores the search Claude ran, the sources it cited and the results it consulted, so you can see which pages it had in front of it.
- Each Claude answer costs 6 credits. Example: 20 questions × 3 runs = 360 credits per sweep. You can switch Claude on or off per brand.
Does Claude recommend you?
Start free and measure Claude alongside ChatGPT, Gemini, Perplexity and the other engines.
Frequently asked questions
Keep reading
AI visibility
Mention rate with a 95% range, share of voice, citations and GEO Score across 9 AI engines.
ChatGPT
How ChatGPT searches, picks and cites brands, and how GeoRush measures your visibility in it.
robots.txt for AI bots
What each AI bot does, which ones to allow so AI can cite you, and a ready-to-copy robots.txt.
Pricing
Plans from $0 with every engine and module; credits explained with worked examples.