On this page (5)
When a worker fails to confirm a check-in within the time window, the procedure should start an escalation: first an attempt to contact the worker, then a notification to the supervisor, and finally a physical check of the post. Each of these steps needs a named person and a time limit, because a procedure in which “someone should check” usually ends with nobody checking.
The most common mistake is a single-stage procedure: the alert goes to one person and that is where it ends. If that person happens to be driving, the mechanism stays as silent as if it didn’t exist.
How escalation works after a missed check-in
The first stage is an automatic attempt at contact. As soon as the grace period runs out, the system sends the worker a reminder. In most cases that is enough, because the person was busy and confirms within a minute. The reminder weeds out false alarms before anyone is pulled away from their work, which is why it does more than anything else to decide whether the staff accept the system.
The second stage is a notification to the duty officer. If there is still silence after the reminder, an alert goes by text message and email to the duty officer and the security manager. The message should include the name of the post and the time of the last confirmation, because the duty officer needs to know whether the silence has lasted three minutes or forty.
The third stage, a physical check of the post, is not automatic and cannot be automated. It has to be written into the procedure: who goes, how quickly, what they do on arrival and whom they notify. The system provides the signal; the decision and the action are down to a person.
Check-in interval, grace period and escalation times
| Parameter | Typical range | What happens when it is set wrong |
|---|---|---|
| Check-in interval | 30–180 min | Too often: people find ways round the procedure. Too rarely: the response comes after the fact |
| Grace period | 3–10 min | Too short: a flood of false alarms and lost trust in the system |
| Time to stage 2 | 1–5 min after the grace period | Too long: you lose the advantage that automation gives |
| Time to stage 3 | set in the written procedure | Not written down = nobody goes |
The most common reason such systems stop working after a quarter is a grace period that is too short. A dozen or so false alarms a week teach the duty officer to ignore alerts, and then they will ignore a real one too. It is better to set the grace period generously, so that the duty officer still responds to every alert.
What stays in the log after a missed check-in
After every missed check-in, TraxSystem keeps the time of the last confirmation, the moment the window was exceeded, the times the notifications were sent, the list of recipients and the moment the silence ended. This is the material people look for afterwards: in an accident investigation, during an inspection and in discussions with the insurer.
Describe the event in the duty log as well, so that its circumstances stay with the same shift.
Calling for help and delivery of notifications
TraxSystem doesn’t summon help or notify the emergency services. It sends a notification to the designated people in your organisation, and that is where its role ends. Calling an ambulance, the fire service or a mobile response patrol is a decision for the person who received the alert.
Sending a notification is also not the same as it being received. The system records that the text message and email were sent, but delivery is not guaranteed. That is why stage 2 should have more than one recipient, and the procedure should state a time after which a lack of response from the duty officer brings in someone else.
Questions and answers
How many people should receive the alert?
At least two, on different devices. A single recipient is a single point of failure for the whole procedure.
Can the system tell “forgot” from “something has happened”?
Not automatically, because the system only sees a missing confirmation. Stage 1 settles it: the reminder explains most cases within a minute, without involving anyone.
Can alerts be muted for a planned break?
Yes, through the post’s schedule. A post with no schedule is treated as off duty and generates no alerts at all. That is convenient for a planned break, but it becomes dangerous when someone forgets to restore the schedule.
See also: Post check-ins: how monitoring a lone worker works in practice · Lone working on site: the employer’s duties and how to meet them