1. Identify what was purchased (ideally from invoice lines).
2. Apply the best available emission factor per activity.
3. Sum to Category 1.
4. Keep a trail from each tonne back to a document and a factor.
For most companies this is the largest Scope 3 category. It is also where spend-based shortcuts hide the most risk.
Per the GHG Protocol Scope 3 Standard, Category 1 is upstream emissions from purchased goods and services.
In:
Raw materials and intermediates
Components and finished goods for resale or use
Professional services, IT/SaaS, marketing, facilities services
Packaging you buy
Usually not in Category 1 (other categories):
Capital goods → Category 2
Fuel- and energy-related activities not in Scope 1/2 → Category 3
Upstream transport you purchase separately → Category 4
Business travel → Category 6
Employee commuting → Category 7
If your GL lumps “other OPEX” into one account, your carbon model will too. Split by what was bought.
Three failure modes show up in almost every first conversation with a group sustainability team:
1. Only spend is left. The ERP kept the amount and the account code. Product names, quantities, and article numbers were stripped.
2. Services are treated like steel. Consulting gets a manufacturing factor, or a single “business services” EEIO row for everything.
3. Supplier PCFs arrive as one number in an email. No method, no boundary, no year. Teams either reject them or accept them blindly.
Searchers look for “purchased goods and services emission factors.” Here is the hierarchy that holds up in assurance conversations:
| Priority | Factor type | Use when | Quality |
|---|---|---|---|
| 1 | Supplier product LCA / EPD / PCF | Exact SKU or close variant match | Highest |
| 2 | Constructed activity-based factor | Product/service identity known; no published EPD | High if documented |
| 3 | Supplier share of corporate footprint | Services (IT, professional) with published company footprint | Medium |
| 4 | Category / proxy average | Close activity class only | Lower |
| 5 | Spend-based EEIO | Last resort for unidentified lines | Lowest; mark as such |
UK note: DEFRA / UK Government conversion factors are strong for energy and travel. For purchased goods they do not replace product-level factors. Spend-based EEIO tables are a different dataset. Do not mix them silently.
Export supplier invoices for the reporting year. Prefer PDFs or structured files that still contain line descriptions, quantities, and suppliers.
Each line becomes one or more emission activities: what was bought, how much, from whom. One invoice often yields several activities (goods + freight surcharge + fees). Typical range: 3–5 activities per invoice.
Misclassification is how consulting inherits a construction factor.
Record method type on every line. When a better supplier PCF arrives next year, you can replace the factor without rebuilding the whole inventory.
Report:
Category 1 total (tCO₂e)
Share by method quality (primary / constructed / proxy / spend)
Top 20 suppliers and top products by emissions
Explicit residual spend-based %
That quality split is what separates a Category 1 figure from a guess.
Goods example: “Monitor Model X, 24 units.” Prefer the manufacturer’s published product footprint for that model. If missing, build from weight, materials, and manufacturing region. Do not use “IT hardware €” if the model is known.
Services example: “Cloud hosting, Feb invoice.” Prefer supplier product or service footprint; else supplier share of footprint; else a documented intensity model. Avoid a generic “computer services” spend factor if the supplier publishes better data.
Opaque project fee: break into materials / transport / on-site energy when the SOW or annex allows. Residual lump sum stays low quality and visible.
CSRD / ESRS E1: expects Scope 3 where material, with transparent methods. Category 1 is material for almost every product company and many services groups.
SBTi: progress on Scope 3 targets needs data that can move when suppliers change. Spend-based Category 1 often cannot show real reductions (price up can look like emissions up).
See also: CSRD: What Your Carbon Data Needs to Survive an Audit and Does Switching to Activity-Based Data Count Toward Your SBTi Target?.
1. Using spend for Category 1 while claiming “activity-based” in the narrative.
2. Double counting supplier PCFs and the same physical purchase again as a generic factor.
3. Ignoring services because EPDs are rare. Services still need a method, just not a fake product EPD.
4. Mixing factor vintages without documenting which year applies.
5. No entity cut in multi-company groups. Group CFOs cannot act on a single blob.
GHG Protocol Scope 3 Standard
GHG Protocol Technical Guidance for Calculating Scope 3 Emissions
UK GHG Conversion Factors
ISO 14067 (product carbon footprint)