Requested page
/pricing
The route the recognized agent requested from your server.
Prompt attribution without false promises
Can you see what people ask ChatGPT before they visit your website?
Not exactly. A crawler request or referral visit does not reveal a person's private ChatGPT prompt. What you can build is a reliable evidence chain showing the questions you monitor, the answers AI produced, the pages agents requested, and the visits that reached your site.
Example server observation
Recognized request · not a conversation transcript
agentChatGPT-User
methodGET
path/pricing
status200
time2026-07-27T14:32:18Z
promptnot available
The direct answer
Request metadata is useful. It is not hidden prompt data.
Server logs describe an HTTP event. Web analytics describe a browser session. AI answer monitoring describes model output. These systems can be compared by time, page, provider, and topic, but they should not be merged into a fictional one-to-one attribution record.
Privacy-safe interpretation
Say “an AI agent requested this page” or “a visitor arrived from ChatGPT.” Do not say “this exact person asked this exact question” unless you collected that information directly and lawfully.
Agent identity
ChatGPT-User
The declared user-agent identity, subject to provider verification and classification.
Request outcome
200 · 14:32 UTC
Whether the page responded successfully and when the request occurred.
Referral visit
chatgpt.com
A separate human browser session may later arrive with ChatGPT referral information.
What you can observe
Requested page
The route the recognized agent requested from your server.
Agent identity
The declared user-agent identity, subject to provider verification and classification.
Request outcome
Whether the page responded successfully and when the request occurred.
Referral visit
A separate human browser session may later arrive with ChatGPT referral information.
What remains unknown
The exact wording of the user question
The full ChatGPT conversation
The identity or account of the person asking
Whether the final answer mentioned your brand
Whether the page was quoted or cited
Whether the user accepted the recommendation
A better measurement model
Reconstruct the journey with four observed layers
No single layer tells the entire story. Together, they show what buyers might ask, what AI actually answered, which pages systems accessed, and what traffic reached your site.
Worked example
Build a finding you can defend
A defensible analysis records the source of every statement. It does not claim that one referral session came from one monitored prompt. It shows converging evidence around a buyer topic.
Conclusion: “Brand Armor AI is increasingly visible for this question family, relevant owned pages are accessible to OpenAI search, and ChatGPT-attributed visitors reached pricing.” That is useful without pretending to know a private prompt.
Question family
Best AI visibility platform for marketing teams
Tracked prompt library
Answer observation
Brand mentioned in 6 of 10 repeated runs
Prompt monitoring
Citation observation
/features cited twice; competitor comparison cited four times
Citation analysis
Server observation
OAI-SearchBot requested /features and /pricing
Crawler activity
Visit observation
Three ChatGPT-attributed sessions landed on /pricing
Web analytics
Build the evidence chain
Monitor representative prompts and real crawler activity without inventing private-user attribution
Brand Armor AI connects prompt results, citations, competitors, requested pages, and recognized AI crawler activity so teams can investigate how AI discovery works with evidence at every layer.
What you can measure
Recurring prompt-level answers across supported AI platforms
Brand mentions, recommendations, competitors, and citations
Server-side recognized crawler requests and requested pages
Content gaps and publishable actions based on observed results
Frequently asked questions
ChatGPT prompt and visitor attribution FAQ
Can Google Analytics show the exact question someone asked ChatGPT?
No. Analytics can identify a visit attributed to ChatGPT when referral or campaign information is available, but it does not receive the private text of the conversation that led to the click.
Can an AI crawler request contain the user prompt?
Normally no. A server request exposes technical request information such as the requested URL, time, method, user agent, and response status. It does not provide the private ChatGPT conversation or reliably reveal the question behind the request.
Does ChatGPT-User activity mean a real person asked about my brand?
It can indicate a user-triggered retrieval action, but the request alone does not identify the person, reveal the prompt, or prove that your brand was mentioned or recommended in the final answer.
How can I learn which questions create AI visibility?
Build and repeatedly run a representative set of category, comparison, problem, product, and branded prompts. Then connect those answer results with citation data, crawler requests, and referral analytics without treating any layer as exact prompt attribution.
Should I add query parameters or personal data to crawler tracking?
No. Crawler measurement should minimize collected data. Brand Armor AI removes URL query strings by default and does not need cookies, authorization headers, or request bodies to classify recognized crawler activity.
Primary references
Crawler identities and platform behavior change. Use provider documentation as the source of truth and review it before changing robots, firewall, or CDN rules.
OpenAI: Publishers and Developers FAQ
Official guidance on OAI-SearchBot access and measuring referral traffic from ChatGPT separately in analytics.
OpenAI: ChatGPT Search
Official explanation of ChatGPT search, citations, and the role of OAI-SearchBot in website availability.
Perplexity crawler documentation
Official distinction between automated PerplexityBot discovery and Perplexity-User actions triggered by users.
Anthropic crawler documentation
Official distinction between Claude training, search, and user-directed retrieval agents.
Related guides
Continue the measurement journey
Crawler intelligence
Does crawler traffic mean a recommendation?
Separate crawler access from citations, mentions, recommendations, human visits, and conversions using an evidence-based measurement ladder.
Read guideMeasurement
ChatGPT referrals vs crawler traffic
Understand the difference between a person arriving from ChatGPT and an AI system requesting a page from your server.
Read guideCrawler intelligence
AI crawler purpose guide
A practical guide to OpenAI, Anthropic, Perplexity, and Google crawler identities, controls, and the evidence each request provides.
Read guidePrompt strategy
How many prompts should you track?
Choose a defensible prompt-set size based on buyer journeys, coverage, model variance, markets, and the decisions your team needs to make.
Read guidePrompt strategy
Branded vs non-branded prompts
Use branded, category, comparison, problem, and local-market prompts for the right measurement job instead of blending incompatible signals.
Read guideMeasurement
How quickly does AI search update?
Understand the separate timelines for crawler access, retrieval, indexing, citations, recommendations, and future model training.
Read guide