> 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/old-analytics/conversation-analytics.md).

# Conversation Analytics

The **Conversation** page shows how every *conversation* in Cekat AI is performing. Use filters such as date, label and inbox to narrow down the data to what you need.

#### Here are the main metrics on this page:

### **A. Total Conversation**

Shown as a *line chart*, this metric tells you how many *conversations* took place in a given period. It counts *unique conversations* per customer, or Monthly Active Users (MAU), so you can see how many customers are talking to you.

### **B. Peak Chat Hour**

Also shown as a *line chart*, this helps you spot your busiest hours, when the most messages come in.

### **C. Returning Rate**

Shown as a *pie chart*, this compares new customers (*first-time customers*) with customers who come back for another *conversation* (*returning users*). It's a good way to measure retention.

### **D. Resolution Rate**

Shown as a *pie chart*, this shows the percentage of *conversations* resolved by the AI compared with those handled by human agents. It helps you judge how well your AI handles customers.

### **E. Conversation Source**

Shown as a *pie chart*, this shows where most of your *conversations* come from, such as Tokopedia, WhatsApp, Messenger and others. It tells you which channels your customers use most.