Prevent Fuel Theft Against Someone Who Is Paying Attention
Fuel waste is a physics problem. Fuel theft is a person, and the person adapts to whatever you install. KO Fleetz combines deterrence at the tank, litre-level detection while the vehicle is still there, and an evidence trail that holds up when somebody denies it.
Fleet and logistics operators run KO Fleetz in production
The loss is small each time, which is exactly why it survives
Theft that takes a whole tank gets noticed and stopped. Theft that survives for years takes a modest amount, regularly, from a vehicle that looks thirsty rather than robbed. It hides inside the natural variance of consumption, and consumption variance is wide enough to hide a great deal. That is the mechanism: the loss is not concealed by cleverness, it is concealed by your tolerance for the ordinary range.
It also takes more forms than most fleets plan for, and the effective ones do not involve a hose at all. A driver and a pump attendant agree that the receipt says sixty litres while the tank receives forty-five, and they split the difference in cash. The card statement is perfectly clean, because the transaction genuinely happened. A depot bowser under-records what it dispenses and the logbook is completed by whoever was holding the nozzle. Jerry cans get filled on the company card at a station that never sees the vehicle. Organised crews with transfer pumps empty parked trucks in minutes at unlit overnight stops, and increasingly drill tanks from underneath rather than opening the cap, because the cap is the part fleets protect. Each of these defeats a different control, which is why a locking fuel cap on its own moves the theft rather than stopping it. KO Fleetz watches the whole chain — the purchase, the delivery into the tank and what leaves it afterwards — because a control that guards one link only relocates the loss to the next.
The cost is not only the diesel. It is the investigation that goes nowhere, the supervisor who now suspects everyone including the honest majority, the vehicle written off as inefficient and sent to the workshop repeatedly for a fault it does not have, and eventually a depot where nobody trusts the fuel numbers. Price your own exposure with the Fuel Theft Savings Calculator before deciding what to spend on stopping it.
Why the monthly reconciliation catches nobody
Most fleets discover fuel loss the same way. Cost per kilometre creeps up, someone runs the numbers, and the gap between fuel purchased and fuel burned is too large to explain away. By then the evidence is weeks old, the trips are finished, the drivers have rotated, and there is nothing left to investigate. The reconciliation was real and the finding was correct and it is entirely useless, because all you can prove is that a loss occurred somewhere across a month and forty vehicles.
Fuel card reconciliation does not close the gap either, and it matters to be precise about why. A card statement proves fuel was bought and paid for. It cannot prove the fuel entered the tank of the vehicle on the receipt, and it says nothing whatsoever about what left the tank afterwards. Both of the most common theft patterns live in exactly the space the card cannot see. So the conversation collapses into an argument about averages: consumption looks high on this route, this vehicle, this shift, but high compared with what? Without tank-level data every theft looks like poor driving and every heavy right foot looks like theft, and the supervisor ends up accusing the wrong person or, more often, nobody. KO Fleetz breaks that stalemate by measuring the tank itself, so consumption stops being the only witness and a short-fill or a siphon separates cleanly from a heavy right foot.
- A card receipt is evidence of a purchase, never evidence of a delivery. Short-filling is invisible to card reconciliation by construction.
- Monthly variance cannot name a night, a place or a person, and a disciplinary process needs all three.
- Depot fuelling is usually recorded by hand by the person dispensing, which asks whoever might be stealing to also be the witness.
- Physical controls move the problem rather than removing it. Lock the cap and the tank gets drilled; watch the depot and the loss relocates to an overnight stop on the route.
- Consumption variance is naturally wide enough to conceal a steady small loss indefinitely, so the theft ends up filed as a maintenance issue.
- Suspicion without evidence poisons a depot. Once accusations land on the wrong drivers, the honest ones stop cooperating and the informal reporting you actually depended on dries up.
How KO Fleetz resolves this
The KO Fleetz modules that fix it
Prevent Fuel Theft 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
Deter, detect, then prove
Step 1: Make the tank a measured object
A calibrated probe reports level every few seconds against the tank's real geometry. Tanks are baffled and saddle-shaped, so a linear reading misreports volume exactly where it matters: near empty, and immediately after a fill.
Step 2: Verify what arrives, not what was billed
KO Fleetz compares every level rise against the card transaction or bowser issue that claims to have caused it. A sixty-litre purchase that delivered forty-five becomes a discrepancy on a named refill, at a named station, on a named shift.
Step 3: Alert while the vehicle is still there
A steep drop with the ignition off, at a location that is not a depot or a station, is not consumption. It raises an event with a coordinate, a timestamp and a volume, and it lands with a supervisor during the same shift rather than in a report.
Step 4: Filter the physics out first
Sloshing on a gradient, thermal expansion through the afternoon and refuelling itself all move the level. Detection accounts for them, because an alert that fires whenever a truck parks on a hill is muted within a week and then it protects nothing.
Step 5: Close each event with an outcome
Every alert resolves as theft, sensor fault or explained, keeping its sensor trace and location. That record is what an HR or legal process needs, and those outcomes tighten the thresholds over time.
What you use
The capabilities that do the work
Drawn from across the platform — this problem is not solved by one module.
Litre-level drop detection
Fires when tank level falls faster than consumption can explain, carrying engine state, speed and position, because any one of those signals alone produces false alarms.
Refill delivery verification
KO Fleetz reconciles each level rise against the volume you were billed for, the only control that catches collusion at the pump where the transaction itself is genuine.
Depot bowser reconciliation
Internal pump issues post to the same vehicle ledger as retail purchases, and a bowser that consistently over-reports becomes visible across many vehicles rather than one.
Location and geofence context
Fuel events carry a coordinate, so a drop at an unscheduled layby at 3am is visibly a different object from one at a known depot pump.
Physical noise filtering
Gradient, sloshing and temperature drift are recognised as artefacts with a characteristic signature — level returns once conditions settle — rather than raised as theft.
Sensor tamper and health alerts
A probe that goes silent, flat-lines or reports an implausible constant raises its own alarm. A quiet sensor is treated as a problem, never as an absence of problems.
Tank and cap lock monitoring
Smart lock state is reported alongside fuel level, so an opening event without a corresponding authorised fill is an event in its own right.
Investigation evidence pack
Each event retains its sensor trace, coordinates, driver assignment and resolution, giving a hearing a documented sequence instead of a manager's suspicion.
In the product
What this looks like in KO Fleetz
What changes
The difference it makes
- A siphon separates cleanly from a heavy right foot
- Drop, not drift
- Delivered litres checked against the litres billed
- Fill verified
- The alert lands with a supervisor, not in a report
- Caught mid-shift
- A sensor trace a disciplinary hearing can stand on
- Evidence, not hunch
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 |
|---|---|
| Unexplained volume as a share of purchases | Litres bought minus litres delivered to tanks minus litres burned, over a full period. Expect a residual, since sensors have tolerance, and track whether it shrinks. A residual that stays flat while alerts are being closed suggests a route you have not found yet. |
| Short-fill discrepancy rate by station and driver | Refills where delivered volume fell materially short of billed volume, grouped by site and by person. A single station appearing across unrelated drivers is a different finding from one driver appearing across unrelated stations. |
| Time from drop to supervisor response | The gap between when KO Fleetz raises the drop event and when a supervisor acknowledges it. This decides whether detection deters anything at all. An alert read the next morning has already failed, no matter how accurate it was. |
| Alert precision | Proportion of theft alerts resolved as genuine rather than sensor fault or explained physics. Track it honestly and publish it. Precision is what keeps supervisors opening alerts, and once it collapses the whole system is decorative. |
| Consumption variance within peer groups | The spread between vehicles doing comparable work. As losses close, the tail should pull in towards the group. A vehicle that stays high after theft is ruled out has a mechanical or behavioural cause, covered on reduce fuel costs. |
| Sensor coverage and health uptime | Share of the fleet reporting usable tank data, and the proportion of time each probe is healthy. Gaps are where theft relocates to, and they will be found faster than you find them. |
Put a number on it before you talk to anyone
The fuel theft savings 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 prevent fuel theft belongs here — the result, in their words.
Frequently asked questions
Consumption is gradual and correlates with the engine running and the vehicle moving. Siphoning is a steep fall in a short window, usually with the ignition off and the vehicle stationary. KO Fleetz evaluates rate of change against engine state, speed and position together, because any single one of those would generate constant false alarms. A drop of forty litres in four minutes from a cold engine in a layby is not a driving style.
Possibly not, and it is much cheaper to find out first. High consumption has ordinary explanations long before it has criminal ones: idling, empty running, driving habits and mechanical drift each inflate a fuel bill without anybody breaking a rule. Start with reduce fuel costs, which addresses litres that burned without doing work. Come here when the arithmetic genuinely does not close, when litres purchased consistently exceed litres delivered plus litres burned and no route, load or fault explains the gap. Installing theft detection on a fleet with an idling problem finds an idling problem, expensively.
Yes. Anyone who tells you their hardware cannot be defeated is not being straight with you. Probes get disconnected, wiring gets cut, and people who steal fuel professionally know what a fuel sensor looks like. This is exactly why device health monitoring carries equal weight with the fuel signal itself: a sensor that stops reporting, returns a frozen constant or reports implausible litres for its tank size raises a hardware alert of its own. The realistic goal is not an untamperable tank. It is that every route to the fuel leaves a trace somewhere, including the route that involves attacking the monitoring.
That is not a promise any software vendor can make, and the honest answer depends on your jurisdiction and your own process. What KO Fleetz provides is an evidentiary record rather than an accusation: a timestamped sensor trace, a coordinate, a volume, the assigned driver and a documented resolution. That is materially stronger than a monthly variance report and is usually enough for an internal disciplinary process. Whether it meets a criminal standard is a question for your legal team, and it will depend heavily on how consistently you followed your own procedure at the time.
Deterrence comes from response time, not from the alert existing. Fuel theft is mostly opportunistic, and opportunism responds to the perceived chance of being caught this time. A depot where an event at 2am produces a phone call at five past develops a reputation within weeks, and the losses move elsewhere. A depot where the same alert is read at 9am develops the opposite reputation just as fast. If you are not going to staff the response, the honest recommendation is to fix the accounting and the physical controls first, and be sceptical of anyone selling you alerts nobody will answer.
Towards whichever control is weakest, which is why we would rather tell you now. Secure the tank and pressure moves to the refill, because collusion at a pump leaves a clean card statement and needs no hose. Cover the retail refills and it moves to the depot bowser, where the record is a logbook. Cover the bowser and it may move to parts, tyres or cargo. This is not an argument against starting. It is an argument for reconciling the whole chain rather than bolting a probe onto the most visible part of it and declaring the problem solved.
Find out whether you have a theft problem or a burn problem
Send us a month of fuel card transactions and one depot's vehicles, and KO Fleetz will tell you which of the two your numbers point at, and whether it is in fact neither.