Data without insight
Telematics systems record speed, acceleration, braking, idling, fuel and location continuously. Most fleets use a fraction of it, typically a vendor's driver score that nobody fully trusts.
The distrust is usually justified. A score comparing a driver on urban multi-drop work against one on trunk routes is comparing different jobs, and drivers know it.
Adjust before you compare
The analysis that produces change is one that accounts for what each driver was actually doing.
- Load weight, which drives fuel consumption directly
- Route type - urban stop-start against motorway
- Gradient along the route, which varies enormously by region
- Weather, particularly wind and temperature
- Vehicle and its age and specification
- Traffic conditions at the time
Once these are accounted for, the remaining variation is closer to genuine driving difference. That is a fair basis for comparison and, importantly, one drivers will accept as fair.
What actually correlates with fuel use
| Factor | Typical influence |
|---|---|
| Excessive idling | Direct, easily measured, easily reduced |
| Harsh acceleration | Significant, and coachable |
| Speed on motorway | Substantial at the top of the range |
| Anticipation and coasting | Real but harder to measure well |
| Route choice | Often larger than driving style |
| Vehicle maintenance | Tyre pressure and servicing matter more than expected |
The last two rows are worth noting because they are not driver behaviour at all. Fleets sometimes run extensive driver coaching programmes while routing and maintenance offer larger, easier gains.
Use it for coaching, not league tables
Published rankings produce gaming and resentment. Drivers learn what the score measures and optimise for it, sometimes in ways that make actual driving worse.
Individual, specific and private feedback works better: this trip had 40 minutes of idling against 12 on comparable trips, here is where. That is actionable in a way a score out of a hundred is not.
Where incentives are used, base them on improvement against a driver's own history rather than on ranking. That rewards everyone who improves rather than only those on the easiest routes.
Be straight about monitoring
Telematics involves monitoring employees, which carries legal obligations and a real effect on trust. Both are better handled openly.
Tell drivers what is collected, what it is used for, and what it will not be used for. Involve them in designing how it is used. Fleets that impose monitoring silently get resistance and data manipulation; those that are transparent generally get cooperation.
A driver score that ignores the route is a score of the route.