Track / Claude Visibility and Mentions
Claude Brand Mentions & Citation TrackerClaude answers some questions from memory and others by searching the web. Botric runs your buyers’ prompts in Claude every week and records whether you’re named, which pages get cited, and which competitors Claude recommends instead.
Grey Aesthetics
Luminary
myIDFi AI
Claude monitoring means running the prompts your buyers ask Claude on a schedule and recording what comes back: whether Claude names your brand, which pages it cites, which competitors it recommends, and the tone it uses.
Claude is harder to check by hand than most engines because it answers in two modes. Without web search, it relies on what it learned in training and shows no sources. With web search, it cites pages inline. A Claude tracker has to capture both.
What Claude learned in training and what it finds on the web today can tell two different stories about the same brand. You need to see both.
When Claude answers without searching, it draws on what it learned during training. It can name you, but it shows no sources and can lag behind your latest product changes.
ClaudeBotWhen Claude searches, it reads current pages and cites them inline. This is where a new or updated page can start earning citations.
Claude-SearchBot, Claude-UserYourBrand is known mainly for product tours: guided walkthroughs that introduce new users to an app. Teams often pair it with a separate checklist tool.
YourBrand is an onboarding platform for B2B SaaS, with in-app checklists, product tours, and activation analytics.yourbrand.com It’s popular with product-led teams.g2.com
A tracker that only counts mentions would score both runs as a win. Only one of them describes the product you sell today.
Add your domain and up to five competitors. Nothing gets installed on your site.
Botric generates prompts across four intents. Edit or replace any of them, or test a new one in the Prompt Playground first.
Botric re-runs each prompt on Claude every week, next to the other engines, and logs mentions, cited URLs, competitors named, and tone on every run.
Content Gap Analysis surfaces missing content, topics, and intent coverage, with priority fixes. Send the top one to the Content Agent to draft the content.
Tracking tells you where you stand. These are the tools that actually move the number.
Content Gap Analysis surfaces the content and topics you haven't covered, where your search intent coverage falls short, and other gaps in what's published, then gives you priority recommendations so you fix the ones that win mentions in the prompts that matter.
Finding a gap doesn't close it. Someone still has to write the page. Content Agent drafts it for you: a comparison page, answer article, or FAQ, built around the exact prompt Claude answered and the source it cited instead of you.
Committing a bad prompt to tracking burns weeks of data on the wrong question. Prompt Playground lets you test a prompt against Claude first, for free, before it goes on your tracked list.
A citation rate on its own doesn't tell you if you're winning or losing. Botric tracks up to 5 competitors on the same prompts, side by side, so every number has something to compare against.
Being named isn't the same as being recommended. A caveat can undo a mention just as easily as a missing citation. Sentiment analysis flags whether Claude's tone toward you is positive, neutral, or qualified, on every run.
You can spend weeks fixing content and still not know if it worked. GEO Score rolls every mention, citation, position, and sentiment change back into one number, so a fix either moves it or it didn't.
Averages hide the story. Split by intent, the same brand can own a category answer and be described wrongly when a buyer asks about it by name.
Category prompts decide whether you're on the shortlist. Claude searched, named YourBrand second, and cited its onboarding page.
| Run | Result |
|---|---|
| Mentioned | Yes · 2nd |
| Cited | yourbrand.com |
| Competitors named | 2 |
| State | Owned answer |
Comparison prompts are often answered from the competitor's own “vs” page. You're in the answer, but Brand A wrote the framing.
| Run | Result |
|---|---|
| Mentioned | Yes |
| Cited | No |
| Cited instead | brand-a.com |
| State | Remembered, not cited |
Direct brand prompts show how Claude describes you by name. When Claude answers from memory, this is where outdated positioning shows up.
| Run | Result |
|---|---|
| Mentioned | Yes · 1st |
| Cited | No sources shown |
| Sentiment | Neutral |
| Described as | “product tours” |
Informational prompts are where your guides get cited without your brand being named. Good content, missed recommendation.
| Run | Result |
|---|---|
| Mentioned | No |
| Cited | yourbrand.com/blog |
| Brand named instead | Brand B |
| State | Silent source |
Claude AEO starts with access, then works on both modes at once: rank where Claude searches, and make your brand’s story consistent wherever Claude learns. Your tracking data shows which one to work on first.
Anthropic runs three crawlers: ClaudeBot for training, Claude-SearchBot and Claude-User for search. You can disallow ClaudeBot and still allow the two search agents, but block the wrong one and Claude can’t find or cite you at all.
# robots.txt User-agent: Claude-SearchBot Allow: / User-agent: Claude-User Allow: /
Brave Search appears on Anthropic’s subprocessor list for web search. A page on Google’s first page can still be buried on Brave, so check both for the prompts you track.
Claude’s memory comes from everything written about you. If your site says “onboarding checklists” and older reviews still say “product tours”, the older story can win.
Claude cites passages, not whole pages. Put a direct answer under a heading phrased like the question, use specific numbers, and name your product in the sentence that answers it.
Full playbook: the Claude optimization guide and the free GEO Foundations course. Crawler behavior source: Anthropic Help Center.
Infinity tracks all five engines Botric covers, including Claude, with 150 prompts and room for five sites.
Luminary AIFrom a brand-new domain to 123+ AI citations across 7 platforms in 90 days.
Grey Aesthetics178+ AI citations across 21 tracked prompts, and 11 position-one Google rankings, in 60 days.
Everything to know before you track your brand in Claude.
A Claude citation is a link to a source page that Claude includes in its answer when it uses web search. Each citation ties a claim to the page it came from. When Claude answers from its training data instead, it usually shows no sources, so Claude can mention your brand without citing you.
Choose the prompts your buyers ask, run them in Claude on a schedule, and log the answer each time: whether you're named, which pages are cited, and which competitors appear. Botric does this automatically, re-running every tracked prompt weekly and recording mentions, cited URLs, competitors named, and sentiment.
Claude answers some prompts from memory and others by searching the web. Answers from memory don't show sources, so you can be named without a citation. It's also why Claude can describe your product the way it looked a year ago.
Claude GEO (generative engine optimization) and Claude AEO (answer engine optimization) describe the same goal: getting Claude to name and cite your brand when buyers ask. In practice that means ranking where Claude searches, keeping your brand story consistent across the web, and writing passages Claude can quote. The Claude optimization guide covers each step.
Anthropic hasn't published a detailed answer, but Brave Search appears on Anthropic's subprocessor list for web search, and independent studies have found strong overlap between the pages Claude cites and Brave's top results. It's worth checking how your key pages rank on Brave for the prompts you track.
It depends on what you want. Blocking ClaudeBot excludes your future content from Anthropic's training data. Blocking Claude-SearchBot or Claude-User affects whether Claude can find and open your pages for search answers. If you want Claude citations, allow Claude-SearchBot and Claude-User at a minimum.
Infinity ($149/month, or $136/month billed annually). It tracks Claude alongside ChatGPT, Perplexity, Gemini, and Google AI Overview, with 150 prompts and a 7-day free trial. Bit and Byte don't include Claude.
Botric re-runs every tracked prompt on Claude weekly, next to the other engines on your plan, so you can compare Claude with the rest and spot changes early.
Yes. Add up to five competitors and Botric records which of them Claude names and which of their pages it cites, prompt by prompt. Content Gap Analysis then shows the content, topics, and search-intent coverage you're missing, with priority recommendations so you close the gaps that matter most.
Content Gap Analysis surfaces the content, topics, and search-intent coverage you're missing, with priority recommendations on which to fix first. Content Agent then drafts the page: a comparison page, answer article, or FAQ, built around the exact prompt Claude answered and the source it cited instead of you.
No. Botric runs prompts on Claude from the outside, so there's no code, pixel, or tag to add.
Different engines give different answers to the same prompt, and a brand that leads in ChatGPT can be missing in Claude. Tracking both shows where you're strong and where a competitor owns the answer. Infinity tracks all five engines Botric covers in one place.
Enter your site to see whether AI engines mention your brand and cite your pages, and where competitors are winning instead.