From WLTP to Telematics: How Real-World Fuel Consumption Gaps Are Reshaping Fleet TCO in 2026

By the editorial team Telematicsjournal · 22 October 2026

Why a TCO based on WLTP is no longer a reliable budget in 2026

A fuel budget calculated from a new vehicle's WLTP specification sheet rests on an assumption that no longer holds in 2026: that laboratory-certified consumption will approximate what your fleet actually consumes on the road. The ICCT (2026), in its study "On the way to 'real-world' CO2 values? Evidence from 2021–2023 on-board fuel consumption monitoring data in the European passenger car market," analysed data transmitted by on-board monitoring devices (OBFCM) from roughly 8 million new cars registered between 2021 and 2023 across the European Union. This volume fundamentally changes the nature of the discussion: we're no longer dealing with a one-off sample, but with a continuous stream of telematics data.

For fleet managers, the consequence is direct. The gap between certified and real-world consumption is not a fixed margin to be tacked onto the TCO. It varies significantly by powertrain type — and based on available data, it continues to widen rather than narrow.

Key takeaways

• For conventional petrol, hybrid and mild-hybrid vehicles, the gap between real and official consumption rose from 18% in 2021 to 19% in 2023, according to the ICCT (2026) — a fuel line underestimated by roughly one-fifth.
• For plug-in hybrids, the gap reached 401% in 2023 (approximately 5 times the official value), up 50% since 2021 (ICCT, 2026): the single most poorly calibrated energy line in any fleet TCO.
• Between 2018 and 2023, official average fleet CO2 fell by 28%, versus only 15% under real-world conditions; for petrol engines alone, real-world emissions fell by just 1% (ICCT, 2026).
• Battery electric vehicles remain outside the OBFCM reporting requirement during the first phase of Euro 7, leaving a blind spot on the true energy cost of electrifying fleets.

Why does the test-to-road gap never truly disappear?

This phenomenon is not new, but it stems from the same structural cause: a laboratory test cycle never fully reproduces real-world driving conditions. The ICCT (2015), in its statement "Real-world vehicle fuel economy gap continues to widen in Europe," established that roughly three-quarters of the gap then measured on the NEDC cycle — the certification protocol used before WLTP was introduced in 2017 — stemmed from manufacturers exploiting loopholes in the test: special tire preparation, fully charged battery before testing, air conditioning switched off. The remaining quarter came from technologies more efficient in the lab than in real use, such as stop-start systems.

Dr Peter Mock of the ICCT summarized the mechanism in 2015: "Roughly three-quarters of the gap between laboratory results and real-world driving is explained by manufacturers exploiting regulatory loopholes."

That year, the average NEDC gap reached 40% (36% for private cars, 45% for company cars), equivalent to roughly €450 in annual excess fuel costs for the average vehicle (ICCT, 2015). This figure should not be compared term-for-term with the WLTP percentages cited later: the test cycle changed in the intervening years, which reduced but did not eliminate the gap.

Beyond test loopholes, factors that cannot be reproduced in the laboratory affect real-world consumption, according to the European Commission (2024): traffic, terrain, road condition, ambient temperature, air conditioning use, onboard electronics and driver behaviour. This is precisely what embedded telematics can now capture, journey by journey, where an official test measures a single scenario.

What is the real scale of the gap by powertrain type?

Available figures do not all tell exactly the same story — which is normal given they come from different measurement methods. Three sources show values that appear contradictory for petrol and diesel vehicles, yet they actually measure different samples and periods.

The European Commission's (2024) figure — roughly 20% gap across a sample of 600,000 cars registered in 2021 — rests on still-partial coverage: as of April 2022, the Commission had received data for only 10.6% of cars and 1.0% of commercial vehicles registered the year before. The ICCT (2026) works with a larger dataset spanning 2021-2023, which explains a slightly different central value but the same upward trend. Self-reported data from Spritmonitor.de, lower at 11-12%, might be explained by selection bias: drivers who voluntarily log their consumption on such platforms may adopt more efficient driving habits on average than a typical fleet — a plausible but unproven hypothesis within the study itself. For a fleet manager, the practical conclusion is not to choose a single figure, but to adopt a reasonable range of 15 to 20% for conventional and classic hybrid powertrains — and to treat PHEVs as a separate category, detailed below.

How to translate this gap into a TCO correction?

Converting a percentage gap into a budget correction requires a straightforward calculation step, provided you spell it out rather than apply it blindly. Here is a generic example, excluding local fuel prices, which illustrates the mechanism only — not a prediction of actual cost for any given vehicle.

Calculation assumption: assume a fuel or energy budget line set at €100 per month based on WLTP-certified consumption, as listed in the vehicle's official TCO.

• For a conventional, hybrid or mild-hybrid vehicle, the gap documented by the ICCT (2026) is 19% in 2023. The expected actual budget then amounts to roughly €119 per month (100 × 1.19).
• For a plug-in hybrid, the gap reaches 401% in 2023 according to the same source, a multiplier close to 5. The expected actual budget then jumps to roughly €500 per month (100 × 5) for the same starting official budget line.

This factor applies only to the fuel or energy line of the TCO, not to the total cost of ownership: depreciation, maintenance and insurance follow distinct logics not addressed in this correction. A fleet manager applying a single coefficient — say 15% — across the entire fleet would severely underestimate the fuel budget for PHEVs and, more modestly, for petrol or diesel vehicles as well.

Embedded telematics data allows you to refine this estimate at the individual vehicle level rather than the average category, capturing actual driving profiles rather than a laboratory scenario. The useful correction in 2026 is therefore not a fixed number, but a differentiated grid by powertrain type, updated as data becomes available.

The plug-in hybrid case: the budget line that really blows out

Plug-in hybrids deserve separate attention because they concentrate, in themselves, the bulk of budget risk for a fleet betting on this transitional powertrain.

For PHEVs registered in 2021, the European Commission (2024) measured real-world emissions averaging 3.5 times higher than certified, roughly 4 litres per 100 km or 100 g CO2/km above the official value. Two years later, the ICCT (2026) observes a gap of 401%, or roughly 5 times the official value, up 50% since 2021.

These two figures do not contradict each other: they document the same widening trend at different points in time, with two distinct sources measuring the same phenomenon over an overlapping period. The gap is widening, not stabilizing.

The underlying mechanism largely hinges on actual electric-mode usage: the certified value assumes frequent charging and heavy battery use, whereas a fleet vehicle that is rarely recharged spends most of its kilometres in combustion mode. For a fleet manager, this means the PHEV correction factor depends directly on the charging policy applied within your fleet, far more than for a conventional vehicle.

And electric fleets: the measurement blind spot

OBFCM devices record fuel or energy consumption and total distance travelled; manufacturers transmit this data annually to the European Commission (2024). This mechanism already works for the conventional and hybrid powertrains analysed above — but not yet, in the same way, for battery electric vehicles.

Battery electric vehicles currently remain exempt from OBFCM reporting requirements during the first phase of Euro 7. The ICCT (2026) recommends extending this obligation from 29 November 2026 onwards, but this is a recommendation, not a confirmed obligation at this stage — check the official Euro 7 texts when updating a fleet budget.

This blind spot does not prevent you from working through the energy line of an electric fleet by other means. According to FleetRabbit (2026), a specialist fleet management blog — a commercial, not institutional, source and should be weighted accordingly — managed charging would reduce peak demand by 25% compared to unmanaged charging, and off-peak scheduling could cut the energy cost per kilometre by up to 50%. These figures do not rest on the same methodological rigour as the OBFCM data analysed by the ICCT or the Commission, but they indicate that for electric vehicles, the TCO correction lever lies less in the gap between real and certified consumption than in managing charging.

Towards continuous correction rather than a fixed figure

The question posed by these gaps is no longer whether a WLTP figure is right or wrong, but how often a fleet budget should be recalculated. A grid by powertrain type, built from OBFCM data available today, offers a more reliable starting point than a single safety coefficient applied across an entire fleet. Yet it remains a snapshot in time: the gaps documented here have grown every year since 2021, with nothing indicating they will stabilize. This dynamic extends beyond consumption alone: between 2018 and 2023, average official CO2 reported by manufacturers fell by 28%, versus only 15% under real-world conditions — and barely 1% for petrol engines alone (ICCT, 2026), a sign the gap affects the stated decarbonization trajectory as much as the fuel budget. One zone remains poorly lit — the true energy cost of electric fleets, for which embedded telematics already has the sensors, but not yet, everywhere, the data collection framework equivalent to that for conventional powertrains.