Skip to content
  • There are no suggestions because the search field is empty.

Agent Analytics FAQs

1. Which plans include Agent Analytics?

Agent Analytics is available on the Standard, Premium and Custom plans. Event limits differ per plan, see the pricing page for the exact figures. The trial period also has access.

2. How do I get my data into Agent Analytics?

You connect a data source under Agent Analytics → Connect a data source. Different connectors are available; pick the one that fits your setup. See here for more information.

3. Which AI engines and agents does it track?

Agent Analytics reports on ChatGPT, Claude, Microsoft Copilot, Perplexity, Google AI Overview, Google Gemini, AI Mode and an "Other" bucket. Agents are grouped into three categories, see more information here:

  • On-Demand AI Fetcher - fetches a page in real time in response to a user's prompt (e.g. ChatGPT-User, Claude-User, Perplexity-User).
  • Search Index - crawls to build or refresh a search index (e.g. PerplexityBot, OAI-SearchBot).
  • AI Training / Data Scraper - crawls to collect training data (e.g. GPTBot, GoogleOther, ClaudeBot).

4. Which data does your agent/worker collect - which logs does it read, and which details does it fetch?

Only AI crawler and agent traffic is collected. Normal visitor/user traffic is filtered out explicitly and never sent to us. From each qualifying request we use the URL that was visited, the user-agent (which identifies the AI engine), and the timestamp; the referer and host are used when present. IP addresses are used only momentarily for de-duplication and are removed before anything is stored.

5.What data, if any, is collected through the logs, and how is it used?

No personal user data is stored. IP addresses are stripped before the data reaches our storage, and we keep no personally identifiable information (PII). What remains is non-personal request metadata, the visited URL, the AI agent's user-agent, the timestamp, and optionally the referer, used solely to attribute AI visits to pages.

6. Do you store my visitors' IP addresses?

No. IP addresses are removed before storage. This applies to every method (file upload, webhook, Cloudflare, Netlify, and the WordPress plugin).

7. Where is the data transfered and under which terms?

Have a look at our security page for further information. The service is used under our standard terms and conditions.

8. Should I turn on consent mode for your agent/worker/plugin? Are you a Data Processor under GDPR?

We are GDPR compliant, but we are not a data processor in the GDPR sense, because we don't process or store any Personal Identifial Information (PII). IP addresses are removed before storage. You don’t need to get the consent from your users.

9. Which file formats and sizes can I upload?

Accepted extensions are .log, .txt, .csv, .json, .ndjson, .gz and .zip. You can upload up to 20 files per upload, max 1 GB per file. You don't need to pick a format, OtterlyAI auto-detects it. See here for more details.

10. Can I mix upload methods? 

Yes, it is possible, log file upload and then the real time connector.

11. Can I upload historical logs?

Yes, that's exactly what File upload is for. Export your web server's access logs for the period you care about and upload them; OtterlyAI extracts every AI agent visit, no live connection required.

12. How long does it take for data to appear in Agent Analytics? 

It depends on the data source, if logfiles are uploaded, the data appears as soon as files are processed. For other sources, as soon as the agent visits the site.

13. After connecting the WordPress plugin, do I see historical data, or only new data from today onward?

The Wordpress plugin and the Webhook and Cloudflare are only monitoring from the date they are activated, but you can additionally upload historical logfiles.

14. Can I connect via Shopify?

Currently not. Shopify does not offer this option yet. See here which upload sources are available.

15. How do I disconnect or remove a data source?

Disabling or deleting the connector in OtterlyAI stops data collection immediately. For Cloudflare, to also remove the deployed components: in your Cloudflare dashboard delete the otterly-agent-<zoneId> Worker under Workers & Pages, then delete the matching route under the zone's Workers Routes.

16. If two brand reports track the same domain, do I need to connect it twice?

No. A connected domain automatically feeds every brand report in the same workspace that uses it. Note that data sources are unique per workspace, though, connections made in one workspace aren't shared with another.