> For the complete documentation index, see [llms.txt](https://docs.cekat.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cekat.ai/en/chat/new-analytics-session-based.md).

# New Analytics (Session-Based)

<figure><img src="https://3693109682-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FZNZlvyOVkZXyd3pz2Cql%2Fuploads%2F7p5nMrEt49GXRNBOgpdk%2FScreenshot%202025-11-13%20at%204.35.42%E2%80%AFPM.png?alt=media&#x26;token=6dbde35d-b8d9-4c3f-bca1-c154c5218262" alt="" width="563"><figcaption></figcaption></figure>

**New Analytics** is a major update to reporting in Cekat AI. Everything is now calculated per **session**, replacing the old model that used the **conversation** as its basis. Every metric, from AI performance and agent performance to SLA, is calculated by **session ID**, covering interactions handled by both AI and agents.

This only applies to **new chats that came in after the feature was released**. Conversations that were already active before the release aren't counted in session-based analytics until they're *resolved*. After that, when the customer gets in touch again, the system starts counting from **Session 1**.

Your old analytics data is still available for reference, but it won't be updated any more. All new data is calculated using the **session-based** approach.

A **session** is created automatically whenever a new conversation starts after the feature is turned on. Sessions are numbered in order for each customer, for example **Session 1**, **Session 2** and so on.

This new approach gives you more accurate, context-rich insight into how well your AI and agents are looking after customers.