Coaching

Detection without a response is monitoring. The response is a conversation, and how it is conducted decides whether the programme works.

What it is

A short conversation about a specific event, with the context.

Soon after — days, not weeks.

Conducted by someone who understands driving, with the clip or the data available to both.

Ending in something specific: an agreement, a piece of training, a change to the round, or a conclusion that nothing was wrong.

What it is not

A performance review.

A disciplinary step, which is a different process with different safeguards.

A monthly ranking delivered by email.

Or a group session about everyone's scores, which humiliates and teaches nothing.

Opening it

With the event, not the score.

"There was a hard brake on the Thursday round, let's look at it" is workable.

"Your score is in the bottom ten percent" is not, because the person's first response is to dispute the ranking and the conversation never reaches the driving.

Ask what happened before showing an opinion. The driver frequently remembers, and the explanation is frequently the end of it.

Listening for the structural cause

A round with an impossible schedule produces late braking.

A junction that everyone brakes hard at is a route finding.

A vehicle with poor mirrors produces a pattern.

Coaching that finds the same cause across several drivers has found a fleet problem, and continuing to coach individuals about it is both futile and resented.

Who does it

Trained, and with time allocated.

Peer coaching works well in some fleets — an experienced driver rather than a manager — and is worth considering where the relationship with management is strained.

And it should not be the person who would conduct a disciplinary, for the same reason as in video review.

Measuring whether it works

Not by the coached drivers' subsequent scores, which improve anyway through regression to the mean.

Against a comparison group, which is the only honest method.

Or by collision and near-miss rates at fleet level over a long enough period, which is slower and more meaningful.

A programme that cannot demonstrate an effect should say so rather than citing score improvements that would have happened regardless.

Open with the event, not the score

The difference between a conversation and a dispute.

"There was a hard brake on the Thursday round, let's look at it" goes somewhere.

"Your score is in the bottom ten percent" produces an argument about the ranking, and the driving is never discussed.

Ask what happened before offering an opinion. The explanation is frequently the end of it.

Connect policy and data

The choices in this note can be compared with project time tracking for safety coaching. Keep the written purpose in control and enable only the information needed at this boundary.

Independent reference

For a thematic point of reference, see the International Organization for Standardization. Use the current material as an outside check before turning the principle into a system rule.