Driver scoring: how it's calculated, and weighted

Short version. Driver scoring turns telematics events (harsh braking, speeding, idling, fuel burnt) into one number per driver, so that a fleet can compare drivers and track them over time. Almost every vendor does it the same way: detect events, divide by distance driven, weight each behaviour, and map the result onto a fixed scale. The differences are in the weights, which decide what the score is really about. Routly’s eco score weights fuel at 40% because fuel is what the fleet pays for; a safety-first fleet would reasonably choose otherwise.

What is driver scoring?

A driver score is a weighted summary of how someone drives, normalised so that drivers doing different amounts of work can be compared. It exists because the raw material is unreadable: a month of a hundred vehicles produces tens of thousands of behavioural events, and nobody reads them. They get exported to a spreadsheet once, discussed in a meeting, and never opened again.

What changes behaviour is a comparison: this driver against that driver, this month against last. That requires three things the events do not have on their own: normalisation, weighting, and a trend. Driver behaviour monitoring produces the events; driver scoring is the layer that makes them comparable.

How driver scores are calculated across the industry

The published methods from large vendors share a skeleton and differ in emphasis.

Geotab scores drivers from 0 to 100 based on rule violations, with each rule given a percentage weight and the weights required to total 100%. Its default scorecard puts excessive speeding at 30%, speeding and seatbelt at 20% each, and harsh acceleration, braking and cornering at 10% each, then groups drivers into risk bands from Low (90–100) to High (0–60).

Motive publishes its formula in some detail. Its Safety Score runs from 50 to 100, computes each behaviour’s event rate per 1,000 miles over a rolling four-week window, deducts weighted points from 100, and shows no personal score until a driver has 100 miles of data. Its defaults weight camera-detected behaviours such as phone use and tailgating most heavily, because the weights are meant to predict collision risk.

Webfleet goes the other way. OptiDrive 360 gives scores from 1 to 10 per indicator, and its indicators include fuel consumption, coasting, gear shifting and constant speed alongside speeding and harsh events.

So the common steps are: count events, normalise by distance, weight, scale. A safety score weights what predicts crashes; an eco score weights what burns fuel. Neither is wrong, but you should know which one you are looking at.

Normalise, or you are just measuring who drives most

The first mistake is ranking by event count. A driver doing 4,000 km a month will out-brake one doing 800 km, and telling them so is telling them they work more.

Everything has to be per unit of exposure: per hundred kilometres, or per hour of driving. Once you do that, the ranking changes completely, and it becomes defensible in the conversation that follows, which is the actual test.

Routly’s eco score: the weights

Routly computes an eco score from 0 to 100 for every trip when the trip ends, and is explicit about what goes into it:

Each component scores linearly between its good and poor ends, and the weighted sum is the trip’s score: 80 and above is green, 60–79 yellow, below 60 red.

Fuel is the largest because it is the outcome, and the rest are largely causes of it. A score dominated by harsh braking measures a driving style; a score dominated by fuel measures the thing the fleet is paying for. That choice is arguable and you should make it deliberately: a fleet whose priority is accident reduction rather than cost would reasonably raise harsh events and lower fuel.

Three implementation details with real consequences:

Not every vehicle reports RPM. When it is absent, its 5% is redistributed proportionally across the other four components rather than treated as a zero. Treating a missing signal as a perfect score flatters drivers of older vehicles; treating it as a failure punishes them. Redistribution is the only version that compares like with like.

Not every vehicle has a fuel sensor. Without usable fuel data, or on a trip too short to measure, the fuel component takes a neutral 75 rather than 0 or 100, for the same reason.

The fuel band is fixed, and tuned for light commercial vehicles. A heavy truck at 30 L/100 km scores zero on fuel however it is driven. If your fleet is mostly heavy vehicles, the fuel component will not separate your drivers, and you should know that before you read the ranking.

“Speeding” means over the limit configured for the vehicle, 90 km/h where none is set. It is not the posted road limit.

Violations get their own ladder

Alongside the eco score, a separate driver score maps violations to a 1–10 scale. Each violation carries penalty points (by default 5 for speeding, 3 for each harsh acceleration, braking or cornering event, 1 for idling), the total is normalised per 100 km where distance is known, and the result goes through a ladder that is deliberately non-linear:

The bands widen as they go, which is the point. The difference between a driver with zero penalty points and one with seventeen is worth a full grade; the difference between 500 and 900 is not, because both are already at the end of the conversation. A linear scale wastes all of its resolution on the tail where nothing changes.

What a score must never be

Three rules, and the third is not optional.

Not a league table on the wall. Public ranking produces gaming: coasting to a stop from further back, avoiding difficult routes, taking the flat way round. You get a better score and worse operations.

Not a comparison across incomparable work. Urban delivery against motorway trunking is not the same job. Compare within a cohort or normalise for route type; a driver who scores badly because their round is all roundabouts has learned only that the system is unfair.

Not an automatic consequence. Behavioural scoring of a named person is personal data with weight of its own, and an automated decision producing legal or similarly significant effects engages Art. 22 of the GDPR. Scoring used for coaching is defensible; scoring that automatically triggers discipline is a compliance problem and a labour-relations one. Human in the loop.

What actually moves the number

In our experience the pattern is consistent and unglamorous:

1. Show the driver their own score, before anyone else discusses it. Most improvement happens here, without a conversation.
2. Talk to the bottom decile, not everyone. A monthly five-minute conversation with four drivers beats a memo to forty.
3. Pick one component at a time. “Your idling is the highest on the fleet” is a thing a person can change this week. “Your score is 62” is not.
4. Show the trend, not the snapshot. Improving from 40 to 60 deserves more recognition than sitting at 70, and recognising it is what keeps the exercise alive past month two.

The score is not the outcome. Fuel consumed per hundred kilometres across the fleet is the outcome, and it is what you should watch to decide whether any of this is working. If scores improve and consumption does not, the score is measuring the wrong thing or people are gaming it; either way the answer is in the fuel number. The KPI board puts those side by side, and the fleet KPIs worth tracking covers the rest of that board.

Frequently asked

How is a driver score calculated?

Count behavioural events per driver, divide by distance driven (Motive uses events per 1,000 miles, Routly per 100 km), weight each behaviour, and map the weighted total onto a fixed scale: 0–100 for Geotab and Routly’s eco score, 50–100 for Motive, 1–10 for Routly’s violation score. The weights are what make one vendor’s score different from another’s.

What is a good driver score?

There is no universal answer, because each vendor uses its own scale and weights. Within a system, the vendor’s bands are the reference: Geotab calls 90–100 low risk, and Routly’s eco score marks 80 and above as green. Across systems, compare trends and rankings, not raw numbers, and do not carry a score from one platform to another.

Driver scoring vs driver behaviour monitoring: what is the difference?

Behaviour monitoring detects and records events: a harsh brake at 14:02, ten minutes over the limit on the A4. Scoring aggregates those events, normalises them by distance and weights them into one comparable number. Monitoring tells you what happened; scoring tells you who to talk to first.

How much driving do I need before a score means anything?

Enough distance for the normalisation to be stable: a few hundred kilometres a month, as a rule of thumb. Motive withholds a personal score until 100 miles. On trips under 0.5 km, where a rate per 100 km would explode, Routly charges a flat penalty per harsh event instead. Scoring a driver on 40 km of data produces noise that will be argued with, correctly.

Can I change the weights?

Yes. Weights live in a scoring profile that can be set for the whole organisation, a group or a single vehicle, and they must be set together and sum to one, so nobody can quietly put one fleet on a different scale. The bands for idling, speeding, harsh events and RPM are configurable the same way; the fuel band is not. Decide once, then leave them alone: weights that change every quarter make trends meaningless, and the trend is the view that drives improvement.

What about accidents rather than style?

Harsh events correlate with risk but do not predict individual accidents well enough to act on. Treat behavioural scoring as a fuel and wear programme with a safety side effect, and be honest about that framing internally. If collision risk is the goal, a camera-based safety score of the Motive kind is the better instrument.

Do drivers have to be told?

Yes. Transparency is an obligation, not a courtesy, and the programme works better when they can see their own numbers anyway.

Can I try it?

The demo has a KPI board with harsh braking by driver on generated data. The weights above are what the software ships with.