> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rankraft.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent Analytics FAQ

> Common questions about crawler tracking, referrals, privacy and data retention.

<AccordionGroup>
  <Accordion title="Why don't I see crawlers with the website snippet?">
    AI crawlers don't run JavaScript, so a browser snippet never sees them. The snippet counts people who arrive from ChatGPT, Perplexity and other AI answers. To see crawlers, use the [Cloudflare Worker](/agent-analytics/integrations/cloudflare-worker), the [WordPress plugin](/agent-analytics/wordpress/introduction) or [server logs](/agent-analytics/integrations/server-logs).
  </Accordion>

  <Accordion title="Will an integration slow down my site?">
    No. The Cloudflare Worker passes every request straight to your site and sends RankRaft AI a copy *after* the response has gone out. The WordPress plugin records the request after the page is sent and delivers logs in the background about once a minute. The snippet loads asynchronously and sends one small request per page view.
  </Accordion>

  <Accordion title="Why does ClaudeBot show 0% verified?">
    Anthropic publishes no IP ranges, so its crawlers can only be matched by user agent. They are shown as **Partial**. See [Crawler verification](/agent-analytics/crawler-verification).
  </Accordion>

  <Accordion title="Why are genuine Googlebot or GPTBot hits marked Spoofed?">
    Your integration is probably reporting a proxy's IP address instead of the visitor's. See [Why genuine crawlers can show as Spoofed](/agent-analytics/crawler-verification#why-genuine-crawlers-can-show-as-spoofed).
  </Accordion>

  <Accordion title="Do you store my visitors' IP addresses?">
    No. IP addresses are used to verify crawlers and are then stored only as a salted hash. The salt rotates every day, so hashes can't be linked across days. Raw user agents, query strings and referrers aren't stored either. Human visits are counted per day and never logged individually.
  </Accordion>

  <Accordion title="How long is data kept?">
    Individual request records (AI crawlers, search bots and AI referrals) are kept for 13 months. Daily totals are kept for the life of the site. The delivery log on Integrations keeps 90 days.
  </Accordion>

  <Accordion title="What does 'AI share of traffic' mean?">
    AI crawler visits plus AI referrals, divided by all requests RankRaft AI counted for your site in the period (humans, bots and AI). Requests to private paths or other hostnames aren't counted.
  </Accordion>

  <Accordion title="Can I exclude checkout or account pages?">
    Yes. Checkout, cart and admin paths are excluded by default. Add more under **Integrations → Advanced → Private paths**. See [Site data controls](/agent-analytics/integrations/site-data-controls).
  </Accordion>

  <Accordion title="What happens to logging if my subscription lapses?">
    Integrations are told the site is paused and stop sending. The WordPress plugin shows a notice in wp-admin with a link to billing. Once billing is fixed, the plugin resumes on its own. If a payment fails, logging continues during the grace period so your trend has no gaps.
  </Accordion>

  <Accordion title="I run a staging site. Will it pollute my data?">
    Only hostnames on the site's **Allowed hostnames** list are accepted. Your domain and its `www` version are added automatically. Data for any other hostname is dropped, so a staging copy or a leaked key can't add traffic to your reports unless you add that hostname.
  </Accordion>
</AccordionGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.