A spend-based estimate gives you a number. It does not give you a basis for decision-making. For procurement teams under CSRD pressure to reduce Scope 3 emissions from their stamped steel supply chains, this distinction matters, because the decisions you make based on a spend-based footprint are likely to be wrong in a systematic, predictable way.
The problem is structural. Spend-based estimates multiply your procurement spend by an industry-average emission factor. That factor represents the average carbon intensity of all steel stamping in your regional economic input-output database, typically the ecoinvent or Exiobase derivation your reporting consultant uses. But your actual stamped steel suppliers are not average. Their energy intensity, energy mix, and production efficiency vary by a factor of three to five. The average tells you almost nothing about which specific suppliers drive your actual emissions.
Why the variance matters more than the average
In a typical mid-market white goods manufacturer’s supply chain, five to eight stamped steel fabricators represent the bulk of Category 1 spend. If you use a spend-based estimate and then try to prioritise reduction efforts, you’re allocating spend as a proxy for emissions. That means your largest supplier by invoice value becomes your apparent biggest emitter, regardless of whether they run efficient gas cogeneration or decade-old resistance furnaces.
The consequences compound. Reduction targets get set against the wrong baseline. Supplier development conversations start from incorrect numbers. The OEM buyer enters negotiations about emissions reduction with a counterpart who knows their own actual energy data and immediately sees the discrepancy. This erodes credibility in a supplier relationship that CSRD has already made more complex.
The specific risk for stamped steel in NACE C25.5
Stamped steel fabricators in NACE C25.5 are one of the supply chain segments where primary data collection is most tractable, and where variance from the industry average is most material:
- Energy intensity range: 180–850 kWh per tonne of stamped output across the segment, depending on press technology, automation level, and facility age
- Energy mix: A supplier using 80% renewable electricity has a fundamentally different carbon intensity than one drawing from a coal-heavy regional grid, but spend-based estimation treats them identically
- Scale effects: Smaller fabricators often show 30–40% higher energy intensity per unit than large facilities, but may represent significant spend in a specialised supply chain
A spend-based estimate for this category can be off by 200–400% in either direction for specific suppliers. That’s not uncertainty you can manage with a caveat in your methodology notes.
What the audit exposure actually looks like
External assurance for CSRD Scope 3 requires auditors to assess whether your methodology produces a “true and fair view” of your emissions. A spend-based estimate for Category 1, when primary data was clearly available and not collected, does not meet that standard for material suppliers.
In practice: if your top-10 stamped steel suppliers represent more than 15% of your total Category 1 footprint, and you have not made a documented effort to collect primary data from them, your auditor has a problem with your disclosure. Not a fatal problem in year one, but a finding, and findings in sustainability reports are increasingly public.
Outline
How spend-based emission factors are constructed
- Economic input-output tables and what they actually measure
- Why industry averages systematically misrepresent specific supplier intensities
- The ecoinvent/Exiobase derivation chain most consultants use
Doing the variance calculation for your supply chain
- A framework for identifying which suppliers differ most from average
- Data sources that give you directional primary data quickly
- When a proxy upgrade (from spend-based to activity-based) is worth the effort
Building the business case for primary data collection
- Cost of estimation error in reduction programme allocation
- Audit and disclosure risk quantified
- Supplier relationship dynamics when primary data enters the conversation
The transition path from spend-based to primary data
- Which suppliers to prioritise first
- What documents to request and why they’re sufficient
- How to update your footprint disclosure mid-cycle
What good supplier primary data looks like for NACE C25.5
- Minimum viable document set: electricity invoices + production logs
- Optional upgrades: gas records, refrigerant data, waste disposal
- What to do when suppliers claim data isn’t available