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

# Post Analysis

> Analyze completed calls for summaries, structured data, or follow-up actions.

# Post Analysis

After a call ends, you can run **post-analysis** on the transcript and metadata to get summaries, structured output, or to trigger integrations.

***

## Overview

Post-analysis runs once a conversation is over. You can configure what kind of analysis or action you want.

There are **three options**:

***

## 1. Call summary

The **Call Summary** option produces a short summary of the conversation.

* Set a **prompt** that tells the system what to emphasize (e.g. issues, resolutions, next steps).
* The output is plain text you can use in notes, emails, or reports.

> **Example prompt:** “Summarize the conversation focusing on customer issues, agent responses, and resolution steps.”

***

## 2. Structured output

**Structured Output** uses a prompt to turn the conversation into **structured data** (e.g. JSON).

* Useful for **reporting, analytics, or automation**.
* The result can be parsed or sent to other APIs.

> **Example:** Extract fields like `intent`, `sentiment`, and `next_action` in JSON.

***

## 3. Integration trigger

This option **calls an integration** when the call ends.

* Send post-analysis data to an external HTTP API.
* Use it for CRM updates, follow-up messages, or logging summaries elsewhere.

***

## Summary

| Option                  | Purpose             | Example              |
| ----------------------- | ------------------- | -------------------- |
| **Call summary**        | Short text summary  | Notes, email reports |
| **Structured output**   | Structured insights | Dashboards, APIs     |
| **Integration trigger** | Call external API   | CRM, logging         |