Why a carrier's emissions number is only half the answer

A freight forwarder routes the same lane through three carriers over a quarter and gets back three emissions figures, each stated with the same confidence. One carrier has never measured its own fuel consumption, so its figure leans entirely on a generic industry default. Another tracks fuel use across its fleet, but only as an average per vehicle type, not per truck. The third monitors fuel consumption on every vehicle it runs, and its figure is built on the most accurate emission factor of the three. All three figures look the same on the page. They are not the same kind of number.

That is the problem a forwarder juggling several carriers actually has to solve, and it is not the one "shipment-level calculation" usually gets credit for solving. Calculating at the shipment level fixes the annual-total blind spot: it makes lane-by-lane and carrier-by-carrier comparison possible in the first place, instead of one blended figure for the whole year. It does not, on its own, say whether the shipment-level figure a forwarder just received from a given carrier is built on that carrier's per-vehicle monitoring, a fleet-level average by vehicle type, or a default the carrier never moved past. That is a separate question, and it is the one that decides whether the number reflects real, improvable performance or is a stale placeholder still waiting to be replaced.

What sits behind a carrier's number

The GLEC Framework, ISO 14083's practical companion, rates the reliability of any transport emissions figure on a data quality tier system running from 1 to 4. Every tier is still a calculation, activity data multiplied by an emission factor, and every tier still has to allocate a vehicle's emissions down to the shipments it carried; nothing measures one shipment's emissions in isolation, a fuel meter reads the vehicle, not the cargo. What changes between tiers is how good the vehicle-level activity data behind that emission factor actually is.

For a forwarder, that scale shows up as three practical states a carrier's own data can be in:

• No primary data — the carrier has not measured its own fuel consumption. Every figure it reports rests on GLEC's network or modal default (Tier 4): a generic, deliberately conservative activity estimate and emission factor for that mode and vehicle category.
• Fleet-level data, by vehicle type — the carrier tracks its overall fuel consumption, broken down by vehicle type (rigid trucks, articulated trucks, vans). This produces a real, carrier-specific emission factor for each type, better than a generic default, but still an average across every vehicle of that type and every trip it ran.
• Per-vehicle monitoring — the carrier tracks fuel consumption on each individual vehicle, including its backhaul rather than only its loaded legs. This is the most accurate emission factor available, calibrated to how that specific vehicle actually performs. It is still not a shipment-specific reading: the fuel a vehicle burns on a trip covers whatever it carried on that trip, so even here the vehicle's emissions still have to be allocated across the shipments on board, by mass or volume share, to produce any one shipment's figure.

Why the same figure at a different tier is not the same figure

Two carriers can return an identical emissions number for the same lane. If one carrier's figure is built on its own per-vehicle monitoring and the other lands on the same value from a network default, both are legitimate: a GLEC default is a deliberately conservative, standards-sanctioned fallback, not a guess. What differs is what each number can do next. The measured figure is anchored in how that specific vehicle actually performs, so it gives the carrier a real, improvable baseline: evidence of where fuel efficiency or routing could bring the number down, and a way to prove it if they do. The default figure cannot move. It was never tied to that carrier's fleet, so it stays the same no matter what the carrier does, stale by design rather than wrong. Under CSRD reasonable assurance review, that is the actual distinction being checked: not whether a default is wrong, but whether a company can demonstrate its own measured, improvable performance instead of resting indefinitely on a fallback.

What a forwarder can actually check

The data quality tier behind a carrier's number is something a forwarder can ask for, not something that has to be assumed. A GLEC Declaration - the standard carbon receipt a carrier issues alongside its figure - states the data quality tier used and the methodology behind it, precisely so a buyer can tell the two apart without re-deriving anything. A carrier that hands over a number with no stated tier is asking to be taken on trust rather than evidence. A stated tier is still the carrier's own claim, though: verifying that claim, rather than just asking for it, is a further question, one a later piece on what makes an emissions record checkable by someone who wasn't there when it was produced will cover in full.

That distinction, tier stated versus tier assumed, is enough to start differentiating a carrier base:

• Which carriers state a data quality tier at all, versus a bare figure with no methodology attached.
• Among those that do, which run per-vehicle monitoring on the vehicles that matter most, versus a fleet-wide average by vehicle type applied everywhere.
• Whether that tier holds steady shipment to shipment, or drifts depending on who filled in the form.

None of this requires auditing a carrier's fleet directly. It requires asking for the data quality tier as a matter of course, the same way a rate or a transit time gets asked for, and treating silence on it as information in itself.

Why it matters for carriers too

A stated, defensible data quality tier is what lets a forwarder compare a carrier favourably against a competitor quoting the same lane. Carriers face their own version of this incentive problem: a network default is not a neutral placeholder, it is GLEC's own conservative, deliberately pessimistic number, and a carrier that never moves past it is handing every comparison to whichever competitor bothered to measure. Tracking fuel by vehicle type is a reasonable next step; per-vehicle monitoring on the vehicles that run the highest-volume lanes is usually where the return is largest, because that is where a forwarder is most likely to be comparing carriers side by side in the first place.

CO2Path records the data quality tier alongside every shipment-level figure it produces, the same way for every carrier and every lane, so a forwarder comparing carriers is comparing tier-labelled numbers rather than bare ones. It does not manufacture per-vehicle fuel data a carrier never measured, no platform can do that, but it makes clear, shipment by shipment, which tier a number actually rests on.

See your carrier base's data quality tiers in one place → Request a demo