A live security desk.
A different way to work.
Agents working alarm dispatch and remote-workforce procedures in our own live control room. Six weeks of recorded results, with the gaps kept visible.
1 September – 14 October 2026 · Our own operation, not an external customer testimonial.
Workforce work items completed
MeasuredWorkforce cases completed without a person
MeasuredRequired workforce check calls followed up
Measured vs operatorThe work.
A security desk handles alarms, dispatch, shift starts, late attendance and required check calls. Each follows an operating procedure, with people responsible for the decisions that need judgement.
What Aptic does.
The Alarm & dispatch and Remote workforce modules bring the procedures, connected tools and human hand-offs together. Work stays with a case, and a recorded outcome is the unit we count.
Basis for this figure
Median over 108 dispatches on the current agent. An earlier dashboard shows 68 s over 121 including the legacy agent; the full-desk review shows median 71 s, P95 140 s. Before: the operator's previous control room, 5 min typically and up to 15.
Basis for this figure
Aptic 98.3–98.4% (98.3% with the test site and placeholders removed). Before: the operator's own 33%.
Basis for this figure
Before: misses on the staffed desk, operator-reported, 7–21 a week.
Completed follow-ups versus the operator’s previous rates
ModelledBasis & assumptions
About 2,086 completed against about 1,030 at the operator's old rates (2.02×). Modelled on those rates.
Calls checked against their own trace
MeasuredBasis & assumptions
Every call run through rule checks; flagged calls read by a person.
Lower, an estimated 18–27% depending on your set-up.
EstimateBasis & assumptions
Depends on desks, volumes and set-up; sized with the prospect during the pilot.
The limits stay visible.
A work item is a case. A call is a channel event; a check call is a required workforce check-in. These results keep those denominators separate. Trace coverage measures calls, not all actions or all work items.
Recorded does not mean flawless. Missed check calls and note promises that reached only the case log remain part of the evidence. No result here is attributed to journey replay or the decision core.
A specific failure, and the rule it led to
An agent cannot tell a caller someone is on the way before a response is actually dispatched. It happened five times early on; a rule now blocks it, and it has happened once since.
The human comparison figures are the operator’s own. The follow-up workload comparison is modelled; cost is an estimate depending on set-up. These are results from one security operation, not a platform-wide promise.
See every figure and its basis ↗What would change
in your operation?
Start with one procedure and an agreed baseline. Measure the result in your own context.