A behaviour score that drivers believe is unfair changes nothing except morale. This is built around driver identification, normalised comparison and scoring you can explain line by line.
| Event | Typically indicates | Common false positive |
|---|---|---|
| Harsh braking | Following too close, distraction | Someone else cutting in — check the footage |
| Harsh acceleration | Aggressive driving, fuel waste | Pulling a heavy load uphill from standstill |
| Harsh cornering | Speed inappropriate for the turn | Roundabouts on a poorly tuned threshold |
| Speeding | Deliberate, or unfamiliar route | Stale road-limit data, or a temporary roadworks limit |
| Excessive idling | Engine left running at stops | Legitimate cab air-conditioning in Gulf summer |
| Night driving | Fatigue exposure | Scheduled night operations — exclude by shift |
The telemetry is the easy part. Accelerometers detect harsh events reliably, speed against a limit is arithmetic, and idling is a timer. Every platform in this market does this adequately. What decides whether a behaviour programme survives is whether drivers accept the numbers, and that is a design question rather than a sensor question.
This is the most common structural mistake. If events attach to a vehicle rather than a person, then any vehicle used by more than one driver produces a score that belongs to nobody. Drivers know this immediately, and the first time someone is questioned about an event from a shift they did not work, the programme loses its credibility permanently.
Identification by iButton or RFID at shift start is inexpensive and it is the foundation everything else rests on. Driving without identification should itself be flagged — not as a disciplinary matter at first, but because unattributed driving silently corrupts every average you calculate.
A driver covering 400 kilometres a day will accumulate more harsh events than one covering 80, regardless of skill. Ranking on raw counts therefore produces a league table of who drove furthest, presented as a table of who drives worst. Normalising per 100 kilometres fixes this, and it is the first thing an experienced driver will point out if you have not done it.
Vehicle class matters equally. A loaded three-axle truck and a panel van do not produce comparable accelerometer readings, so thresholds are set per vehicle type and comparisons are made within groups. Comparing across classes is how you end up with a scorecard that the truck drivers correctly dismiss.
Idle-time scoring imported unmodified from a European fleet will flag nearly every driver in the region during summer, because leaving the engine running to keep a cab habitable at 45°C is not misconduct. Either exclude ambient-driven idling from scoring, or set a threshold that reflects reality — and if the objective is genuinely fuel reduction, the answer is cab cooling equipment rather than a score that punishes drivers for the climate.
Fleets that see behaviour change coach the bottom group and reward improvement. Fleets that see behaviour hidden use scores punitively, and drivers respond by learning to game the thresholds — coasting to avoid braking events, accelerating gently then speeding. The data is identical in both cases; the outcome is not.
Publishing the scoring method to drivers before the programme starts costs nothing and removes the most common objection, which is that the numbers are arbitrary. Reviewing disputed events with route playback, and with video evidence where it exists, settles the rest.
A behaviour programme is worth running only if drivers accept it. Book a demo or message us on WhatsApp — tell us your fleet mix and we will show you a scorecard built on your vehicle classes rather than a generic one.
Driver identification by iButton or RFID at the start of a shift, so events attach to a person rather than a vehicle. Without it you are scoring vehicles, and any vehicle with more than one driver produces numbers that belong to nobody. Unidentified driving is flagged separately so it does not silently corrupt the averages.
Reliable at detecting the event, less reliable at explaining it. A harsh brake caused by another vehicle cutting in looks identical in accelerometer data to one caused by following too closely. This is why disputed events should be reviewed with route playback, and why fleets running dashcams resolve behaviour disputes considerably faster.
Not the way a European template scores it. Leaving an engine running to keep a cab habitable at 45°C is not misconduct, and a threshold that ignores this will flag nearly every driver and destroy the programme’s credibility. Either exclude ambient-driven idling or raise the threshold to something defensible — and if fuel is the real concern, cab cooling equipment addresses it better than a score.
That is your call rather than ours, but the pattern we see is that linking scores directly to pay produces gaming rather than improvement — drivers learn the thresholds and optimise around them. Coaching the lowest group and rewarding measured improvement tends to change actual behaviour. The data supports either approach; only one of them usually works.
The scores stabilise within a few weeks once identification is in place and thresholds are tuned. Behaviour change follows the coaching, not the measurement — fleets that publish the method, review disputes fairly and coach consistently tend to see movement within a quarter. Fleets that install it and say nothing see none.
A demo against your vehicles and your numbers is more useful than a sample dataset. Tell us your fleet mix and we will set it up.
Book a Live Demo →