Monitoring Overview
Review every call your agents handle: outcomes, recordings, transcripts and captured fields, with Observability watching the platform underneath.
Showing instructions for Xebit VoiceCloud.Showing instructions for Xebit VoiceKit, on your own infrastructure. Switch with the selector at the top. Updated 22 October 2026.
Every call an agent handles leaves a full record: a recording, a turn by turn transcript, a summary, an outcome and the structured fields you defined with the agent. Monitoring is where your team reviews those records day to day, call by call and agent by agent, and decides what to follow up. It sits on top of Observability, which watches the platform itself and opens incidents when something breaks.
Monitoring and Observability
Both look at the same calls, but they answer different questions.
| Monitoring | Observability | |
|---|---|---|
| The question | What happened on our calls, and what should we do next? | Is every layer of the platform working, and if not, why not? |
| Who uses it | Operations, sales, collections and support teams | Engineers and whoever is on call |
| What you look at | Calls, outcomes, recordings, transcripts, captured fields and agents | Per call and per node metrics, event logs, replay and incidents |
| What it does by itself | Nothing. Your team reviews and acts. | An SRE agent opens incidents, writes the root cause analysis and proposes a fix that an engineer approves |
When one call looks wrong, open its replay to hear the recording with the transcript and event log side by side. When many calls look wrong at once, check Observability for an open incident before you start changing agents. See Observability and the Observability overview.
Where to find it
- Sign in to the Xebit VoiceCloud dashboard at
voice.xebit.aiand open the calls list. Every inbound and outbound call across your agents appears there. - Sign in to the Xebit VoiceKit admin console on the host your team deployed it on and open the calls list. Every inbound and outbound call across your agents appears there.
- Results for a single campaign live inside that campaign. See Campaign Analytics.
What every call leaves
| Record | What it gives you |
|---|---|
| Recording | The whole call, every leg, including anyone who joined from a browser |
| Transcript | Turn by turn, in the language the call happened in |
| Summary | What the caller wanted, what was agreed and what happens next |
| Outcome | The call's disposition, such as completed, failed to connect or caller hung up |
| Structured fields | The fields you defined with the agent, such as a lead status, a promise to pay date or a booked slot |
| Event log | What the platform did during the call, shown beside the transcript when you replay it |
For more on how these records are produced, see Conversational Intelligence.
Where the data goes
You do not need to copy anything out of a call. The record can reach your systems as the call ends:
- Webhooks fire on call lifecycle events, outcomes and transfers. See Outbound Webhooks.
- The REST API lets your systems pull calls, outcomes, recordings and transcripts when they need them. See the API Reference.
- The Zoho and Salesforce connector writes the outcome and your fields to the right record. See Integrations.
Xebit does not send WhatsApp, SMS or email to your customers. If you want a message after a call, a webhook fires and your own messaging system sends it. Incident notices from Observability are the exception: they can reach your own team by email, Slack or Teams.
On an air gapped Xebit VoiceKit deployment nothing leaves the network, so only systems reachable inside it can receive webhooks or connector updates.
How long data is kept
On Xebit VoiceCloud, recordings, transcripts and outcomes are stored in Xebit's infrastructure. How long they are kept is set out in the data retention policy.
On Xebit VoiceKit, recordings, transcripts and outcomes are kept in your own storage. Retention is whatever your team configures under your own policy; Xebit does not set it.
In this section
- Call Analytics: call volumes, outcomes, duration and filters.
- Agent Performance: how each agent is doing and how to improve it.
- Call Recordings: listen to any call.
- Transcripts: read what was said, with the summary and captured fields.
- Call Signals: intent, urgency and follow up reasons as fields you define.
- Lead Quality Metrics: lead status across your pipeline.
- Conversion Tracking: how many calls led to a result.
- Export Reports: send call data to your own tools.
Next step
Start with Call Analytics to see the calls list and the filters you will use every day.