The Thing You Are Spending

Every part of a lone worker system depends on people using it as intended, and every part of it is easy to work around silently.

When the thing you are spending also depends on reliable work records, employee tracking software can support time, attendance and workload review without being treated as the emergency response itself. Compare any workflow with HSE lone-working guidance and keep alarm ownership, escalation and dispatch responsibilities explicit.

What cooperation looks like

Carrying the device and wearing it where it works.

Checking in accurately rather than out of habit.

Reporting a near miss at the end of a shift that is already over.

Withdrawing from a situation and saying so.

Telling somebody the device failed rather than living with it.

All five are voluntary, and all five are what make the system describe reality.

What spends it

A response that was slow and never explained.

A check-in time used in a conversation about lateness.

A device that tracks continuously when people were told it did not.

A withdrawal questioned afterwards.

A near-miss report that produced nothing.

Each is individually small. The effect is cumulative and it does not return quickly.

How the loss appears

Not as a complaint.

As a falling near-miss rate, which reads as an improvement.

As check-ins that arrive exactly on time and mean nothing.

As devices reporting from vehicles.

As withdrawals ceasing.

And as an alert log that looks healthier every quarter while describing less, which is the specific danger: the measurements improve as the system empties.

The asymmetry

Trust is slow to build and fast to lose, and its loss is invisible in exactly the data used to judge the programme.

Which is why the reporting and withdrawal figures matter beyond their face value: they are the only early indicators available, and they move before anything else does.

What builds it

Publishing the response times, including the poor ones.

Publishing the false alarm rate, so accidental activations are unremarkable.

Acting visibly on the first reports.

Making the first withdrawal obviously costless.

And keeping the stated purpose, which costs nothing until the week somebody wants the data for something else.

The single most effective thing

Somebody withdraws, and nothing happens to them, and everybody hears about it.

No policy statement achieves what that one event does, and an organisation that has never had one either has no risk or has a policy nobody believes.

The honest tension

Sometimes a worker is genuinely not using the system, and that matters.

But the first question is why, and the answer is usually comfort, a belief about tracking, or an experience of the system failing.

Treating it as a compliance matter before asking produces the behaviour it was meant to correct, which is the pattern this whole note describes.

The measure

Near-miss reports and withdrawals, both of which should be non-zero and neither of which should be falling.

And the question asked directly once a year: do you believe somebody is at the other end?

The answer is the programme, and it cannot be read from any log.

Rebuilding it after a breach

One incident of the data being used for something else is enough, and pretending otherwise wastes the recovery.

Say what happened and that it was wrong.

Change the configuration so it cannot recur, and show the change.

Then wait, because nothing else works quickly.

An organisation that restates its policy without changing anything has confirmed the workforce's reading.

The annual question

Asked directly, of a sample, without a manager present.

Do you carry it? Do you believe somebody is watching? Has anything you were told turned out not to be true?

Three questions, ten minutes, and the answers are the only honest measure of the programme's standing.

The small things that build it

Replying to a near-miss report within a day, even to say nothing will change and why.

Fixing a device that somebody said was uncomfortable, rather than explaining that it meets the specification.

Telling a worker the outcome of an alarm they raised.

And publishing the response time in the month it was poor.

None of these is a programme feature. Together they are the programme, because they are what people use to decide whether the system is for them or about them.