For many companies, Scope 3, the emissions in their value chain, is the largest part of their footprint. It's also the part they can't measure directly, because it happens in someone else's factory, truck, or farm.
I've seen a version of this problem in agricultural supply chains through EUDR work: getting reliable data from many upstream actors is the bottleneck, not the calculation.
Why it's hard
- The data belongs to others. Suppliers may not track emissions at all.
- Many suppliers. Large companies may have thousands.
- Inconsistent methods. Suppliers who do report may use different boundaries and factors.
- Allocation. A supplier's total emissions must be split across its customers.
- Confidentiality. Suppliers may not want to share operational data.
- Deep chains. Tier 1 suppliers depend on tier 2 and 3.
The three main methods
The GHG Protocol's Scope 3 guidance describes a ladder of methods:
| Method | How | Accuracy | Effort |
|---|---|---|---|
| Spend-based | € spent × factor per € for that sector | Low | Low |
| Average-data | Physical quantity (kg, km) × average factor | Medium | Medium |
| Supplier-specific | Supplier's own product-level emissions | High | High |
Most companies use a hybrid: spend-based for the long tail, better methods for the biggest categories.
A practical improvement loop
- Baseline with spend data from your ERP. Fast and complete.
- Find the hotspots. Usually a few categories and suppliers dominate.
- Collect activity data for hotspots: quantities, materials, transport distances.
- Engage top suppliers for specific data, with a simple template.
- Score data quality and show it next to every number.
- Repeat yearly, moving more emissions up the ladder.
The software side
What makes Scope 3 tooling useful:
- Supplier portals that are dead simple. A form beats a 40-page questionnaire.
- Data quality scores per line item.
- Versioning so you can explain year-over-year changes ("did emissions fall, or did our method change?").
- Mapping tables from ERP spend categories to emission factor sectors.
- Evidence attachments for audit.
The data model overlaps a lot with traceability systems. See Data Models for Agricultural Supply-Chain Traceability.
Common mistakes
- Treating spend-based numbers as precise.
- Chasing every supplier equally instead of the hotspots.
- Ignoring categories because they're hard (e.g. use of sold products).
- Reporting improvements that are actually method changes.