Trust infrastructure lets someone outside your company rely on a freight emissions number without trusting only your word. It has four parts. First, the same method every time. Second, a record that links the number to what produced it. Third, a calculation that nobody can change without leaving a trace. Fourth, a way for others to check all of this themselves.
A freight forwarder gives a customer's auditor a freight emissions figure for one shipment. A carrier with 60 trucks supplied the figure, and it was calculated correctly under ISO 14083, the international standard for freight emissions. Months later, the auditor asks a harder question: how do you know that this number has not changed since it was calculated? The forwarder asks the carrier. The carrier still has the figure, but it has no answer to the question.
A few years ago, the main question in freight emissions was whether anyone could produce a number at all. ISO 14083 and the GLEC Framework, the industry guide to applying it, have answered this. They define the calculation structure: the formula, the allocation rules and a default emission factor for companies that have no better data of their own. An emission factor is the amount of emissions per unit of activity, for example per litre of fuel.
But neither of them guarantees two things. First, that a carrier derived its own measured emission factor carefully. Second, that the resulting figure has not changed since the calculation. This is a gap in tools, not a gap in standards. Closing it needs a different kind of infrastructure than the calculation itself.
This matters as much for a small or mid-sized freight forwarder without a sustainability team as it does for a large company. It may matter more, because your customer's auditor comes to you when a carrier's number is not accepted.
Most freight forwarders who read this will not file a CSRD report themselves, because CSRD, the EU sustainability reporting law, applies mostly to larger companies. But your shipper customers who do file a CSRD report will expect the same standard of proof from you, by contract. This is true even if you never take part in an assurance review yourself. An assurance review is an independent check of the report.
You may have already started to ask carriers which data quality tier they use. A data quality tier is a level that shows how detailed a carrier's data is. Asking is the right first step, but it is not the whole answer. A stated tier tells you what a carrier claims. It does not tell you whether the number can survive a check by someone who was not there when the number was produced. This article is about that second question.
This problem is real. A European food importer published a sustainability report with real CO2 figures per route, for road compared with sea shipping. The report says that its carbon measurement is still at an early stage, and it names no calculation standard behind the numbers. This is not a rare case: today it is closer to the norm than to the exception. It is exactly the gap that trust infrastructure is meant to close.
What "trust infrastructure" actually means
For logistics emissions data, trust infrastructure has four parts:
1: The same calculation, every time
CO2Path runs the same formula on every shipment. The formula multiplies activity data by an emission factor, and it divides the result between everything that the vehicle carried. One shipment should never look more trustworthy than another because of a difference in the calculation. Only a real difference in the data behind the shipments should create such a difference.
2: A trail back to what produced it
Every figure should be linked to what produced it. The record should show which shipment, which vehicle, which route, which fuel and which emission factor were used. It should also show how the total of that vehicle was divided between the shipments it carried, which is called allocation. Without this trail, a number in a CSRD report is only a claim, because nobody outside the company that produced the number can check it. For you, this is the difference between giving a shipper customer a number and giving it an answer. The customer will ask sooner or later: how do you know?
3: Nobody can change it afterward without leaving a trace
After a calculation is finished, nobody should be able to edit it without leaving a trace. If a figure really needs a correction, the correction should be a new record next to the original, with a reason. It should not silently replace the original. For a forwarder, this matters most long after the shipment, because an audit can take place eighteen months after the shipment moved. By then the carrier may have changed its systems or its owner, or it may no longer answer your calls. A record that exists only in an email thread does not survive this time gap. A record that was locked on the day of the calculation does.
4: Someone outside the company can check it
An auditor, a supply chain partner or an assurance provider should be able to check the method and follow the trail. They can confirm that nothing has changed, and they do not have to rely on the word of the reporting company. This lets you say more than "my carrier told me so" when your own customer asks.
Why your customers ask for more than the regulation requires
CSRD requires an independent check of sustainability reports by a third party. The required level is limited assurance, which is a lighter review. An EU proposal would have raised this step by step to reasonable assurance, a stricter check that is closer to a financial audit. But a 2026 Omnibus amendment removed this increase from CSRD completely. The legal minimum therefore stays at limited assurance for the foreseeable future. This lowers the bar less than it seems, because your shipper customers can ask for more than the law requires.
| What limited assurance checks | What a shipper's own contractual request checks | |
|---|---|---|
| What gets checked | Whether the figure looks plausible | Whether the figure comes from real activity data, whether the same method was used throughout and whether anything has changed since |
| Strictness | A lighter review | Comparable to a financial audit, by choice and not by law |
| What passes | A number with an explanation behind it | A number with a documented record behind it that has not been changed |
Companies that already have this in place can answer such a request immediately. Companies that do not have it will find that rebuilding the records afterwards, for thousands of shipments, is not realistic. For this reason it is worth building the records before the request arrives, and not after it.
In 2025, PwC published guidance on the quality of environmental, social and governance (ESG) data. The guidance tells reporting companies to expect exactly this kind of review. Controls must be documented, not only applied. Any calculated or aggregated figure should receive a second, independent check before the company reports it.
The same guidance also covers missing data. Sometimes a data point is not available from every partner in the chain, so it is normal practice to fill the gap with an estimate. The guidance calls this "frequently the case with Scope 3 GHG emissions". These are the indirect emissions of greenhouse gases (GHG) along a company's supply chain. An estimate is not a warning sign by itself.
Why supply chain data exchange amplifies the requirement
Trust infrastructure matters even more when emissions data moves from one company to another. One shipment usually crosses two company borders: from carrier to forwarder, then from forwarder to shipper. The shipper typically asks first, because it needs a defensible number for its own CSRD report. The request then reaches the forwarder. The forwarder can only pass on what the carrier provided, and it cannot check the number independently.
A carrier that says "trust me, it is correct" does not make a number trustworthy, and neither does a forwarder that repeats this message upward. A number is trustworthy when it comes with its own proof. The proof is a documented method, a traceable history and a way to check the number. This check does not depend on anyone's word at any point in the chain. This is what "self-attesting" means. The data carries its own evidence, so nobody has to accept it on trust at each handover.
Perhaps you are a carrier or forwarder that supplies this data, and you do not report under CSRD yourself. The problem is still yours, from the other side. Every customer that runs a serious audit will sooner or later ask the question at the start of this article. By then "trust me" will not be good enough. You can build the answer once, so that it works for every customer who asks. This costs much less than rebuilding the answer for each customer under audit pressure.
What this looks like in CO2Path
Return to the 60-truck carrier from the opening. Suppose this carrier is on the fleet-level tier: it tracks fuel by vehicle type, not by individual truck. CO2Path takes this emission factor and the activity data for one specific shipment (route, weight and vehicle type). It runs the calculation in exactly the same way as for every other shipment of this carrier. It also runs it in the same way for the shipments of every other carrier, whatever tier their emission factor comes from.
CO2Path locks the result on its ledger together with all the inputs. Months later, the auditor asks whether the figure still matches what was first produced. The carrier does not need to search through emails or an old spreadsheet. The record already gives the answer, and it has not changed since the day of the calculation.
CO2Path does not check whether this emission factor was the best one available. Whether the carrier should have used a more precise factor is a question for the carrier, not for the platform.
But CO2Path always completes the calculation. Sometimes a carrier has no primary data of its own. Primary data is data measured in the carrier's own operations. Then CO2Path uses the GLEC default emission factor for that mode and vehicle category. It labels this value clearly as a default and does not present it as something better. CO2Path handles the calculation in the same way for real data and for a default, at whatever data quality tier. The formula is the same, the record of the inputs is complete, and nothing changes afterwards without a trace.
For a forwarder, the practical next step is simple. Ask which tier a carrier is on. Then ask to see the record behind it, for example the fuel logs or the telematics export that the claim is based on. CO2Path cannot produce this evidence for a carrier. It also never leaves a shipment uncalculated because the evidence is missing.
Asking is only the first half, and what you do with the answer matters just as much. The answer may stay in an email attachment, or it may be locked in a record that nobody can edit without leaving a trace. This is often the difference between a number that you can defend and a number that you must defend in a hurry. The difference is largest when the shipment is eighteen months old and the carrier has a new owner or no longer answers the phone.
None of this requires all your carriers to be sophisticated already. The 60-truck carrier in this example does not run advanced software. It has a fleet-level tier and is willing to say so. You can ask any carrier today which tier it is on, whatever tools the carrier uses. The answer becomes more precise as the carrier's own data improves. This improvement does not have to happen first.
See which of your carriers' records are locked and traceable, and which are not, in one place → Request a demo