What off-hook minutes tells you about server usage

Off-hook minutes in the Server Performance Report is the total number of minutes that all Asterisk channels on the server were off-hook and carrying a call during the selected shift window, giving you a direct measure of how hard the server worked rather than how many calls it processed.

What off-hook minutes counts

Every time a Channel transitions from idle to off-hook — meaning it is actively carrying audio, whether the far end has answered or the call is still ringing — the timer starts. When the channel goes back on-hook, the timer stops and the elapsed seconds accumulate into the off-hook total for the shift. The report then converts the accumulated seconds into minutes for the summary display.

This figure is a utilization number, not a quality or outcome number. It does not tell you how many calls were answered, how many were dropped, or how long agents talked. It simply tells you how many total minutes the server's audio fabric was occupied during the window.

How off-hook minutes differs from call count

The summary block also shows the number of calls the server handled during the shift. Off-hook minutes and call count tell different stories. A shift with many short calls — typical of Predictive dialing at an aggressive Dial level with high abandons — produces a high call count but a lower off-hook total than a shift where the same number of calls stayed connected much longer. Off-hook minutes reflects the intensity of audio work more accurately than call count alone.

A quick way to derive the average call duration from the report: divide off-hook minutes by the call count. If that implied average is very short, the server was handling a lot of rapid-fire attempts rather than sustained conversations. If it is long, the channels were tied up in extended calls, which keeps the Concurrent calls figure high throughout the shift.

Using off-hook minutes for capacity planning

Off-hook minutes is most useful when you track it across shifts of similar length and comparable agent counts. A consistent upward trend in off-hook minutes over several weeks means your call volume is growing. Pair that trend with the system load figures from the same shifts: if load is also rising proportionally, you are approaching the point where adding more agents or campaign lines will push the box past its comfortable operating range.

flowchart TD
  A["Compare off-hook minutes across shifts"] --> B{"Trend rising?"}
  B -->|Yes| C["Call volume or duration increasing"]
  C --> D["Check system load trend"]
  D --> E{"Load also rising?"}
  E -->|Yes| F["Approaching capacity limit"]
  F --> G["Plan resize or second server"]
  E -->|No| H["Volume up but box handling it"]
  H --> I["Continue monitoring weekly"]
  B -->|No| J["Stable utilization"]
  J --> K["Baseline established - use for forecasting"]

A useful benchmark: take the off-hook minutes from your busiest shift this month and compare it against the same shift last month. If the number has grown by more than 20–25 % without a corresponding reduction in average system load, the box is carrying more audio work with the same or less headroom — that gap is where degradation begins.

What off-hook minutes does not tell you

Off-hook minutes is purely a server-level figure. It does not map to agent talk time — those numbers live in the campaign and Agent performance reports, which filter by campaign outcome. It does not tell you about call quality, Latency, or whether audio was clear on any particular call. It tells you how heavily the server's channels were occupied, nothing more.

Off-hook minutes sits alongside system load, Asterisk channel count, and CPU percentages in the same summary block. For how all those figures relate to each other and what to do when they signal a problem, see the guide to monitoring VICIdial server health and capacity.

For a step-by-step walkthrough of the full report including how to read the graph portion, see how to read the Server Performance Report graphs.

Tracking off-hook minutes manually across shifts takes discipline. On a VICIfast managed server, performance logging is on from the first boot so your baseline data starts accumulating the moment your box is live. Start a VICIfast trial and your server is running in under 40 seconds.