Improve Driver Safety With Evidence Your Drivers Will Accept
Safety programmes rarely fail on the technology. They fail the first time a driver is shown a league table he can dismantle in one sentence. KO Fleetz measures behaviour against the road that was actually being driven, so the conversation starts with a fact instead of an argument about whose route is harder.
Fleet and logistics operators run KO Fleetz in production
The collision is the cheap part
A crash generates a bill that arrives in pieces over eighteen months, and only the first piece looks like a crash. There is the excess, the repair, the third-party claim, and then the long tail: a vehicle off the road for weeks waiting on panel work, a load moved at spot rate, a driver signed off, an investigation that consumes days of management time, and a premium at renewal repriced against your record rather than your intentions. The event lasted two seconds. The cost runs for two years.
The larger leak has no incident report attached at all. Hard acceleration burns fuel that produced no distance. Late, heavy braking turns pads into a consumable measured in months instead of years, and destroys the discs behind them. Aggressive cornering and kerbing shred tyre shoulders. Riding a clutch in traffic writes off a component whose replacement takes a vehicle out of service for a day. None of that is a safety incident. All of it is the same behaviour, surfacing in the maintenance budget where nobody connects it back. KO Fleetz connects it back, reading heavy braking and hard acceleration as the pads and the fuel they spend, not only as events on a scorecard.
Then there is the driver you cannot replace. A safety culture that reads as surveillance costs you the experienced people first, because they are the ones with somewhere else to go. Recruiting and running in a replacement is expensive in ways that are easy to underestimate. The free driver cost calculator is a reasonable way to work out what an hour of driver time and a departure genuinely cost you, before you design a programme that risks causing them.
You cannot manage safety from the incidents you noticed
Most fleets run safety off the incident report, which is a lagging indicator with a selection bias built in. It counts the events that hit something hard enough to be worth reporting. It does not count the ones that missed. A driver having four genuinely terrifying near misses a month and no collisions has a clean record and a serious problem, and the record cannot tell him apart from a careful driver on the same route. You are managing a sample that excludes almost all of the risk.
The alternatives are worse. The sticker on the back door invites the public to report your drivers, which produces anecdote filtered through whoever was annoyed enough to dial. The annual toolbox talk delivers general advice to a room, months away from any specific thing anyone did. And the first instinct once telematics arrives, ranking drivers by raw event counts, is the fastest way to lose the room, because it puts the multi-drop city driver at the bottom every single week for the crime of driving in a city.
- Near misses leave no trace. The event that mattered most this month is the one where nothing happened, and nothing is exactly what got recorded.
- Raw harsh-braking counts measure traffic density, not driving. Comparing a motorway trunker to a multi-drop van is comparing two different jobs.
- Absolute speed thresholds are close to meaningless. Sixty is fine on one road and a serious offence on another two hundred metres later.
- Coaching that arrives weeks after the event is coaching about an abstraction. The driver does not remember the roundabout and cannot learn from it.
- Safety-critical defects reported at a walk-around go into the same backlog as a broken cab light, so a brake fault queues behind cosmetics.
- Fatigue is invisible in the incident record. The relevant fact is usually the shift shape that preceded the event, and nobody was looking at rosters and driving hours on one screen.
How KO Fleetz resolves this
The KO Fleetz modules that fix it
Improve Driver Safety is not one feature — it is a few KO Fleetz modules working off the same records. These are the ones that carry it.
The workflow
Normalise first, coach second, measure third
Step 1: Capture the event with its context
KO Fleetz stores a harsh braking event with speed, position, road class, time of day and what the vehicle was doing in the thirty seconds either side. Without that context it is a number, and a number is arguable.
Step 2: Normalise against exposure
KO Fleetz rates events per distance driven and groups them by road type and conditions, so a driver is compared against people doing comparable work rather than against a fleet average that describes nobody.
Step 3: Score, then show the working
The scorecard is published with the comparison group visible underneath it. A driver who can inspect the basis of his score argues about his driving. A driver who cannot argues about the software.
Step 4: Coach the specific segment
Replay the trip to the exact moment. A conversation about a named junction on a named Tuesday changes behaviour in a way that a monthly score never has.
Step 5: Check whether it actually worked
Track the coached behaviour for that driver afterwards. A programme that never verifies its own effect is a reporting exercise wearing a safety badge.
What you use
The capabilities that do the work
Drawn from across the platform — this problem is not solved by one module.
Contextual event capture
Harsh braking, acceleration, cornering and speeding are recorded with the road class, position and speed history around them, not as a bare tally.
Exposure-normalised scorecards
KO Fleetz rates behaviour per distance within a peer group of comparable work, which is the only version of a driver ranking that survives contact with a driver.
Speeding against the posted limit
Measured as time and distance spent over the limit for the road being driven, rather than against one blunt fleet-wide threshold that is wrong nearly everywhere.
Segment replay for coaching
Open the trip at the second the event fired and see the approach. This is what turns a coaching session from an accusation into a debrief.
Zone-based speed rules
Depots, customer sites, school zones and yards carry their own limits, so a walking-pace rule inside a warehouse yard is enforceable rather than aspirational.
Safety-critical defect escalation
Brake, steering, tyre and lighting defects raised at inspection are separated from cosmetic faults and routed to stop the vehicle rather than to join a queue.
Shift shape and driving hours
Continuous driving time and shift patterns sit alongside the event record, because the useful question after an incident is usually what the eleven hours before it looked like.
Coaching record per driver
KO Fleetz holds what was discussed, when, about which event, and what the rate did afterwards. This is both a management tool and the evidence trail your insurer will ask for.
In the product
What this looks like in KO Fleetz
What changes
The difference it makes
- Behaviour rated against comparable roads and work
- Like for like
- Coaching lands on a named junction, not a score
- Replayable
- The event rate after a session is tracked, not assumed
- Coaching checked
- Brake and steering defects routed to stop the vehicle
- Faults first
How you will know it worked
Measure these
We publish no savings figure for your fleet, because we have not measured your fleet. These are the metrics that will move if this works.
| Metric | How to measure it |
|---|---|
| Harsh event rate per distance, by road class | Count events divided by distance, split by road type. Never report the raw count. The raw count measures the route, and using it will cost you the credibility of the whole programme in about a fortnight. |
| Time spent over the posted limit | Sum the minutes above the limit for the road being driven, as a share of driving time. Time beats event count here, because one sustained thirty-minute breach and one brief overtake are not the same risk. |
| The spread of scores, not the average | Track the gap between your safest and least safe quartile within a peer group. Fleet average safety scores move slowly and hide everything. The tail is where your claims come from. |
| Repeat-event persistence after coaching | For each coached driver, compare the event rate in the weeks before and after the session. This is the only measure that tells you whether coaching is working or merely happening. |
| Safety-critical defect rate at inspection | Count brake, steering, tyre and lighting defects per inspection, by depot. A vehicle that is mechanically unsafe is a safety failure no amount of driver coaching addresses. |
| Brake and tyre replacement intervals | Track consumable life by driver where vehicles are consistently assigned. It is a slow signal and a very honest one, because it cannot be gamed by driving carefully for a fortnight. |
Put a number on it before you talk to anyone
The driver cost calculator is free, shows its formula, and asks nothing of you. Work out your own figure, then bring it to a demo.
Customer evidence
Fleets using KO Fleetz to do this
Customer story pending. A short, attributed quote from a fleet that used KO Fleetz for improve driver safety belongs here — the result, in their words.
Frequently asked questions
Yes. Camera Intelligence runs computer vision over dashcam, in-cab and bay feeds, clips the events telematics alone cannot see — distraction, a following distance that collapses, a phone raised to the ear — and files each as a short clip tied to the vehicle, driver and moment. It sits alongside the telematics-based safety scoring rather than replacing it: motion data tells you something happened, the clip tells you what. The one rule we hold to is that a person reviews and confirms every flag; the model surfaces candidates, it does not penalise a driver on its own.
Not entirely, and pretending otherwise is how programmes fail. You are measuring how people do their job, in detail, all day. The distinction that matters to a driver is what happens to the data. If it is used to coach a specific event and then to notice the rate improved, most drivers come round, particularly the good ones who have privately known who the risky driver is for years. If it feeds a monthly ranking that feeds a disciplinary process, they were right and you will lose the room. The tooling does not settle this. Your policy does.
We will not promise that, because the honest chain of causation has three links and we only supply one. Software identifies the behaviour. A supervisor conducts the coaching. The driver changes what they do. Break any link and nothing improves, and the link that breaks most often is the second one, because coaching is time-consuming and nobody's week has room for it. Fleets that succeed here allocate the coaching hours before they buy the software. Fleets that skip that step buy a very detailed record of accidents they still have.
By never comparing across different kinds of work. KO Fleetz rates a driver against others driving comparable roads at comparable times, and shows the comparison group alongside the score so anybody can check it. Where a driver has no peer group, they are measured against their own history instead. This matters more than any other design decision on the page: the fastest way to destroy a safety programme is to hand the depot's most experienced multi-drop driver a report saying he is the worst in the fleet.
KO Fleetz tracks vehicles, not people, and the distinction has practical consequences. If a vehicle goes home with a driver, its position is still reported unless you configure otherwise, which is a legitimate concern rather than a technicality. Privacy modes can suppress position outside working hours for vehicles where that is appropriate. The right approach is to decide the policy openly with your drivers, configure it to match, and say so. Discovering the setting exists after somebody has already been asked about their Sunday is a bad way to build trust.
Some are, and a system claiming otherwise is not being straight with you. A pothole, a badly-sited speed bump and an emergency stop for a cyclist all generate deceleration. The mitigations are context and patience: thresholds are calibrated per vehicle class because a laden truck and a van decelerate very differently, each event carries the road and speed around it so an obvious cause is visible, and a location generating events for everyone who drives it gets recognised as a road problem rather than a driver problem. Individual events are noisy. Rates over weeks are not.
See what a fair driver scorecard looks like
Give us one depot and a month of data. We will show you the ranking raw counts produce and the ranking exposure-normalised rates produce, side by side.