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

# Conversations

> Review every call — recordings, transcripts, outcomes, and extracted data

**Conversations** is the log of every call your agents have handled. It's where you verify agent behaviour, audit individual calls, and export outcomes.

## Finding calls

The call log defaults to the last 7 days and filters by:

| Filter                        | Values                                                            |
| ----------------------------- | ----------------------------------------------------------------- |
| **Date range**                | Any period                                                        |
| **Status**                    | `picked`, `not_picked`, `failed`, `busy`, `voicemail`             |
| **Call type**                 | inbound / outbound                                                |
| **Agent / Campaign / Upload** | Narrow to one agent, one campaign, or one specific contact upload |
| **Tags**                      | Tags applied by [call analysis](/platform/call-analysis)          |
| **Phone number**              | Find all calls with one contact                                   |
| **Web vs phone**              | Separate browser test calls from real phone calls                 |
| **Duration**                  | Under 30s, 30–60s, 1–5 min, over 5 min                            |

**Export CSV** downloads the filtered result set (optionally emailing copies); large exports are prepared in the background and delivered by email.

## Inside a call

Click any call to open its details — three tabs:

### Transcript

The full conversation, turn by turn, with timestamps — agent on one side, caller on the other. The header shows **who ended the call** (agent or caller) and the **end reason** (caller hung up, agent completed the conversation, silence timeout, duration limit reached, voicemail detected, and so on). End reasons are the fastest diagnostic when calls aren't going the way you expect.

### Recording

An audio player for the call, with download buttons for the recording and a JSON session report. Listening to even a handful of recordings catches things transcripts hide — pronunciation problems, awkward pauses, talking over the caller.

### Analysis

The output of [call analysis](/platform/call-analysis) for this call: the extracted data fields, the structured output, and the **integration logs** showing whether webhooks for this call delivered successfully (with status and timestamps).

## Reviewing chat sessions

Text conversations have their own log under **Conversations → Chat**, covered in [Chat agents](/platform/chat-agents#reviewing-and-supervising-sessions).

<Tip>
  Make a habit of filtering by status weekly: a rising `failed` or `voicemail` share usually points at list quality or calling-window problems, not the agent. Combine the duration filter with `picked` to find suspiciously short answered calls — they're usually instant hang-ups worth listening to.
</Tip>