The Data Behind the Number: Why Primary Supplier Data Decides Whether Your Product Footprint Survives Scrutiny
In our last article we argued that a product carbon footprint must be governed and traceable, a data object with provenance rather than a number in a spreadsheet. That raises the harder question, the one every CFO asks the moment they accept the premise. If a footprint is only as credible as the data beneath it, where does that data come from, and why is so much of it not good enough? This is the problem that decides whether a published footprint holds up or quietly falls apart under a verifier's questions.
Two kinds of data, two very different levels of trust
Every input to a product footprint is one of two things. It is primary data, measured at the source, the actual electricity a factory drew to make your component, the real quantity of material a supplier consumed. Or it is secondary data, an industry average pulled from a database, a proxy for an activity nobody measured directly. Both standards that matter, ISO 14067 and the GHG Protocol Product Standard, insist you distinguish between them, because the gap in reliability is large. Primary data reflects what actually happened in your supply chain. Secondary data reflects what typically happens somewhere, on average, often in a different country and a different decade.
The uncomfortable truth is that most product footprints today lean heavily on secondary data. It is the path of least resistance. Databases are available on a subscription, a supplier survey is not. So the average number goes in, the calculation runs, and a footprint appears that looks every bit as precise as one built on measured data. The decimal places are identical. The confidence behind them is not.
Why the averages fail you exactly when it matters
A footprint built on generic averages is fine until someone with authority asks it a question it cannot answer. That moment is arriving faster than most companies expect. Under CBAM, a European importer who cannot prove the actual embedded carbon of a product pays a default charge calculated on punitive assumptions, so a secondary-data footprint is not just less accurate, it is more expensive. A customer running their own Scope 3 inventory will increasingly refuse a supplier number that rests on a database proxy, because their own assurance depends on yours. And a verifier assessing an ISO 14067 declaration treats a primary-data input and a secondary-data input very differently, because the standard tells them to.
So the average number carries a hidden liability. It reads as a measurement when it is really an estimate, and the moment it is challenged, by a border official, a procurement team, or an auditor, there is nothing underneath it to defend. Companies discover the weakness of their data at the worst possible time, when the footprint is already published and the challenge is already on the table.
The data-quality ladder, and why direction matters more than position
The answer is not to replace every average overnight, which is neither possible nor necessary. It is to know exactly where each input sits on the ladder from proxy to measured, and to climb it deliberately where it counts. A first footprint will lean on secondary data in many places, and that is an honest starting point, provided the proxies are visible as proxies rather than hidden as if they were measured. What separates a credible programme from a box-ticking one is movement. Each cycle, the inputs that matter most, the ones that make up the largest share of the footprint, shift from estimated to supplier-specific, and the confidence of the whole number rises with them.
This is where the governance we described previously becomes the mechanism rather than the theory. If every input is already graded for quality and tagged with its source, you can see at a glance which parts of a footprint are soft and which are solid. That map is what tells you where to spend. You do not chase primary data everywhere. You chase it where a secondary estimate sits on top of a material contribution, because that is where better data changes the answer and where a challenge would do the most damage.
From passive calculation to active supply-chain engagement
Climbing the ladder means going out and getting the data, and that is a different discipline from running a calculation. It means identifying the suppliers and materials behind your largest emission contributions, then asking them, specifically, for the measured figures the proxy is standing in for. Not a blanket survey, which produces little, but a targeted request tied to the lines that matter most to the number. Over successive cycles, the footprint moves from a model of your product to a measurement of it, and the declaration you publish carries primary data precisely where scrutiny concentrates.
This is the work the next wave of product carbon accounting is really about. The first wave made the calculation fast. The governance layer made it traceable. The data layer, the one most companies have barely started, is what makes the number true. A footprint earns its authority not from the sophistication of the tool that produced it but from the quality of the data that went in, and the honesty with which the gaps are shown. That is the difference between a number that impresses in a slide and one that survives a verifier, a customer, and a border.
At CorpStage, ESG 360 is built to make that climb visible and manageable, so every footprint shows not only its value but the quality of the evidence beneath it, and every data request moves a soft input one rung closer to defensible. The number is the easy part. The data is the business.