Product Carbon Footprints: Reliable Supplier Data

Buyers increasingly ask suppliers for a product carbon footprint (PCF) rather than relying on a single corporate emissions number, but a PCF on its own proves little. A reliable figure is one you can trace back to its system boundary, its data sources, and the share of those sources that came from the supplier itself rather than an industry average, not one that simply arrives with a low number attached. ISO, the GHG Protocol and the World Business Council for Sustainable Development’s Partnership for Carbon Transparency (PACT) are now converging on how to make that traceability standard practice rather than a point of difference between suppliers 1.

This matters commercially as well as technically. CBAM reporting, the EU Battery Regulation and early Digital Product Passport pilots are starting to ask for product-level emissions data that a corporate footprint cannot supply, and suppliers who can produce a defensible PCF quickly are becoming easier to work with than those who cannot. This guide sets out what “reliable” means in practice, which standards now apply, and how to build a data request that suppliers can actually answer.

What “reliable” actually means for a PCF

A PCF is not reliable simply because it is low, precise-looking, or produced by a well-known supplier. Reliability comes from four things being visible alongside the number: the system boundary (cradle-to-gate is the common starting point for business-to-business exchange), the methodology and emission-factor database used, the share of primary versus secondary data behind the figure, and whether it has been verified by anyone other than the supplier itself. A single summary number with none of this detail cannot be compared with a competitor’s figure, checked for double counting, or defended if a customer or auditor asks how it was built.

This is also why two PCFs for similar products can legitimately differ by a wide margin without either being wrong. Different cradle-to-gate boundaries, allocation choices and emission-factor vintages all move the result, so the question for a buyer is not “which number is lower” but “which number is better supported.”

The standards are converging, not competing

For years, buyers and suppliers have had to reconcile ISO 14067 2, the international standard for quantifying a product’s carbon footprint, with the GHG Protocol’s Product Life Cycle Accounting and Reporting Standard, which covers similar ground with its own terminology 3. That reconciliation work is now underway directly rather than being left to individual companies: in April 2026, ISO and the GHG Protocol finalised a joint working group to build a single, harmonised product-level accounting standard rather than maintain two overlapping ones 4.

Alongside this, the WBCSD-convened PACT initiative has focused specifically on how PCF data is exchanged between companies, rather than just how it is calculated. PACT Methodology Version 3, soft-launched in April 2025 after a two-year consultation, is built around the principle that comparability and transparency matter more than forcing every supplier to the same precision on day one 5. For buyers, the practical effect is that a supplier citing ISO 14067, the GHG Protocol Product Standard or PACT is working from a credible base; the detail to check is how they applied it, not which label they used.

Primary versus secondary data: how much do you actually need

There is no fixed percentage of primary data that makes a PCF acceptable, and suppliers who claim otherwise are usually overstating the precision of their own figure. The more useful question is whether primary data covers the process steps that actually drive the footprint. Automotive’s Catena-X network, which runs one of the largest live PCF data-exchange systems in production, makes this explicit by publishing a “primary data share” alongside every PCF value rather than setting a pass/fail threshold, so a buyer can judge the figure on its composition rather than treat all PCFs as equally precise 6.

The direction of travel in corporate-level guidance points the same way. The GHG Protocol’s Scope 3 Standard revision, still in development as of March 2026, would require companies to disaggregate reported emissions by data type and disclose verification status, and it would restrict allocating one corporate-level emissions figure across multiple suppliers to cases where those suppliers are genuinely homogeneous 7. Applied to product-level data, the same logic holds: an emission factor borrowed from a database is reasonable for a low-impact input, but for the highest-emitting steps in a product’s life cycle, buyers should be asking for supplier-specific data and should expect that share to increase over time rather than stay fixed.

Where supplier PCF data most often breaks down

A handful of problems account for most of the PCF data that does not survive scrutiny:

• Boundary mismatches. A cradle-to-gate figure compared against a cradle-to-grave figure from a different supplier is not a like-for-like comparison, even if both are methodologically sound.
• Undisclosed allocation. Emissions from a shared facility or shared input divided across products using an assumption the supplier has not stated cannot be checked or reproduced.
• Corporate-to-product allocation. Taking a site- or company-level emissions total and dividing it by output volume to create a “product carbon footprint” produces a number that looks product-specific but is not; the GHG Protocol’s draft revision would explicitly limit this practice to genuinely comparable production lines 7.
• Stale or unstated emission-factor vintage. Secondary data ages, and a figure built on a five-year-old factor set without disclosure can understate or overstate current performance.
• No verification trail. A figure the supplier calculated and reported without any independent check carries more weight if that is stated plainly than if it is presented as if it were assured.

None of these mean the data is unusable. They mean it needs to be read with the same scepticism a finance team would apply to an unaudited number, and used accordingly in supplier comparisons and decisions.

Building a data request suppliers can actually answer

The fastest way to get better PCF data is to ask for the right things the first time, rather than iterating through several incomplete submissions. A workable request asks suppliers to state: the system boundary applied; the standard or methodology followed (ISO 14067, the GHG Protocol Product Standard, or PACT); the primary data share and which process steps it covers; the emission-factor database and its edition or year; and whether the figure has been verified, and by whom. Our supplier carbon assessment checklist sets out the fuller version of this list for first-time supplier engagement, and the approach holds whether the supplier is in Europe or, as is often the case for buyers with Asian supply chains, further down a multi-tier chain where data systems vary widely by tier and geography.

Treat the first response as a starting point rather than a final answer. Where a supplier cannot yet provide primary data for a high-impact step, agree what they would need to collect it and over what period, rather than either accepting a secondary-data estimate indefinitely or disqualifying the supplier outright.

What is pulling PCF data from optional to commercially necessary

Several regulatory and customer-driven pressures are converging on the same requirement: product-level emissions data that can withstand scrutiny. The EU Battery Regulation already required carbon footprint declarations for electric vehicle batteries from February 2025, with a digital battery passport due to follow from February 2027, making PCF data a market-access requirement rather than a disclosure exercise for that sector 8. CBAM’s own embedded-emissions reporting runs on a related but distinct methodology, and its reliance on regulatory default values is being phased down in favour of actual data from installations, which has knock-on implications for how EU importers request emissions data from Asian suppliers. Wider Digital Product Passport requirements under the EU’s product policy framework are following a similar pattern, starting with specific product categories before broadening.

None of this means every supplier needs a fully verified, primary-data PCF today. It does mean that the gap between suppliers who can produce credible, comparable product carbon footprint data and those who cannot is becoming a commercial differentiator, not just a sustainability one — and that buyers who ask precise questions now will spend less time unpicking unreliable numbers later.

If your team is fielding inconsistent PCF submissions from suppliers, or needs a data request that suppliers can realistically complete, talk to our team about building one that fits your supply base.