How to calculate scope 3 emissions: from spend data to supplier data

Knowing how to calculate scope 3 emissions requires understanding four methods, used in increasing order of accuracy:

• Spend-based (procurement spend × industry emission factors)
• Activity-based (physical quantities × specific emission factors)
• Supplier-specific (primary emissions data from your supply chain)
• Product-specific (product carbon footprints supported by life cycle assessments)

Most companies start with spend-based because it uses financial data they already hold. For companies beginning to measure scope 3 for the first time, this is exactly the right approach. Spend-based is a practical and legitimate starting point, but it is a starting line, not a finish line. This guide covers all four methods, when to use each, and what the progression looks like in practice.

For a broader introduction to why scope 3 matters and which of the 15 categories apply to your business, read our complete guide to scope 3 emissions first.

The four ways to calculate scope 3 emissions

Understanding the four scope 3 calculation methods and when to apply each is the foundation of a credible inventory. The GHG Protocol maps these to four tiers of data quality, from spend-based estimates at Tier 4 through to product-level carbon footprints at Tier 1. Most companies use all four simultaneously, spend-based for the long tail of smaller categories, activity-based where physical data exists, and supplier- or product-specific data where it matters most.

Method 1: Spend-based calculations

Spend-based calculation multiplies procurement spend by an industry-average emission factor drawn from an environmentally-extended input-output (EIO) database, typically Exiobase. The emission factor is determined by the supplier’s industry category and country of purchase.

When to use it: Year one; limited data availability; the long tail of smaller, less material suppliers; categories where physical data is not accessible.

Pros: Fast to implement; uses financial data already in your ERP or accounting system; can produce a first estimate across all applicable scope 3 categories from a single spend file upload. 

Cons: Least accurate. Can overstate or understate actual emissions because it applies industry averages regardless of individual supplier carbon intensity. For example, in a recent engagement with a higher education organization, the carbon emissions calculations delivered by Normative were 72% lower than those conducted using the HESCET tool (which relies on a spend-based approach) for the same academic year. Businesses should also be aware that this approach does not capture whether a supplier is more or less carbon-efficient than their sector peers.

Important nuance: The GHG Protocol recommends an inflation adjustment when using spend data, to ensure emissions do not rise simply because prices rise. For companies using manual spreadsheets, this step is frequently missed, with accuracy implications of up to 20-30% (based on Bank of England Rate using Consumer Price Index) over a multi-year period. 

Calculation example: £1M spend on logistics services × 0.5 kg CO₂e/£ = 500 tonnes CO₂e. A valid baseline estimate, but one that assumes every pound spent on logistics produces the same emissions, regardless of the actual mode, route, or efficiency of the provider.

The goal is not to eliminate spend-based data from your inventory. One of our customers with a mature scope 3 programme, still used spend data for 56% of its inventory in 2025. The goal is to replace it where it matters most.

Method 2: Activity-based calculations

Activity-based calculation uses physical quantities: kilograms of material; kilometres of freight; kilowatt-hours of energy; multiplied by material-specific or activity-specific emission factors. It still draws on average emission factors, but applies them to physical quantities rather than financial proxies, making results significantly more accurate.

When to use it: When you have operational data beyond invoices, such as supplier weight data, logistics reports with distance and mode, and energy metering. It is commonly used for Category 1 (purchased goods), Category 4 (upstream transport), and Category 6 (business travel). 

Pros: More accurate than spend-based; reflects actual physical usage; does not fluctuate with price inflation; enables stable year-on-year comparison.

Cons: Requires data from sources beyond your finance system; may need suppliers to share material weights or transport documentation.

Calculation example: 50,000 km of road freight × 0.8 kg CO₂e/km = 40 tonnes CO₂e. The same freight movement calculated through spend data may produce a materially different figure depending on fuel prices in that reporting year.

A result that surprises many companies: Switching from spend-based to activity-based data can cause reported emissions to change materially (they can decrease, or increase if spend-based was underestimating). This is not a sign of worse performance, it is better accounting. One of our retail customers saw reported scope 3 emissions rise by 30% when shifting to activity-based data. The emissions were already there; the spend-based method had underestimated them. When this happens though, it is important to explain the methodology change clearly in your sustainability report.

Technical note: When activity data is added for a category, the corresponding spend must be removed from the spend-based calculation to prevent double counting the same emissions. 

Method 3: Supplier-specific data

Supplier-specific calculation uses primary emissions data reported directly by your suppliers. This comes in two practical forms: entity-level data (the supplier’s own scope 1, 2, and 3 emissions from their sustainability report, allocated to your share of their revenue), and activity data supplied directly by the supplier (material weights, transport distances, or energy volumes from invoices and delivery records). 

When to use it: For your top 10-20 suppliers by emissions exposure; for categories covered by Science Based Targets initiative (SBTi) reduction targets; when the Corporate Sustainability Reporting Directive (CSRD) requires disclosure of your primary versus secondary data share.

Pros: Reflects actual supplier emissions rather than averages; captures year-on-year improvements as suppliers decarbonize – their progress flows directly into your reported figures.

Cons: Depends on supplier willingness and reporting capability; typically takes one to two reporting cycles to establish meaningful coverage.

Calculation example (entity-level): Supplier A reports 2,000 tonnes CO₂e across scope 1, 2, and 3. Your purchases represent 5% of their revenue. Allocated share: 100 tonnes, compared to a spend-based estimate that might have produced 300 tonnes for the same transaction.

This is also relevant for mandatory reporting and target-setting frameworks. Under CSRD’s ESRS E1, in-scope companies must disclose what proportion of their scope 3 inventory is primary data (supplier-specific and activity-based) versus secondary data (spend-based). For SBTi, near-term targets must cover at least 67% of scope 3 emissions. Both frameworks point in the same direction: a planned, progressive shift from averages toward primary data.

Method 4: Product-specific data (PCF)

Product-specific calculation uses a Product Carbon Footprint (PCF), a quantified measure of the lifecycle greenhouse gas emissions of a specific product, typically supported by a Life Cycle Assessment (LCA) or published as an Environmental Product Declaration (EPD) for construction materials. This is the GHG Protocol’s Tier 1: the most granular and accurate data type available for Scope 3 Category 1. 

When to use it: When you need product-level precision for your highest-emitting purchased goods; when a supplier manufactures multiple product lines with significantly different carbon intensities; or when your own products are subject to PCF disclosure requirements under the Carbon Border Adjustment Mechanism (CBAM), CSRD, or the EU Digital Product Passport (applicable from 2027). 

Pros: The most accurate method available; captures product-level variation that entity-level supplier data cannot; enables direct comparison between suppliers of the same material; when connected natively to your corporate carbon footprint, product-level improvements flow directly into your Scope 3 Category 1 figures without manual reconciliation. 

Cons: Most resource-intensive to produce or collect; requires suppliers to have conducted a full LCA; coverage remains limited in many supply chains, though regulatory pressure is accelerating disclosure.

Calculation example: Your steel supplier provides two grades. Entity-level data gives one average intensity figure across all purchases. With PCF data, you know Grade X emits 1.8 kg CO₂e/kg and Grade Y emits 2.6 kg CO₂e/kg, enabling procurement decisions that reduce actual emissions, not just emissions estimates.

Which calculation method should you use?

The answer for most companies is a hybrid approach. All four methods can run simultaneously, with spend-based covering the long tail and more accurate methods applied where emissions are most significant.

The top 20% of suppliers by spend typically account for 60-80% of procurement-related emissions. Targeting those suppliers first, rather than trying to improve methodology across all categories at once, is where most companies gain the most ground. Hitachi Rail demonstrates what systematic progression can achieve: starting with scope 3 accounting for just 10-13% of its actual emissions footprint, the business reached over 90% coverage by working through its supplier base with Normative.

Want to plan your approach to carbon accounting in more detail? Check out our carbon accounting software buyer guide.

Common scope 3 calculation challenges

Most companies encounter the same three obstacles when improving scope 3 data quality. Understanding them in advance significantly reduces the time spent resolving them.

Reported emissions rising when data quality improves. One of the most consistent findings across our customer programmes: switching from spend-based to activity or supplier-specific data often increases reported emissions, because actual emissions are higher than industry averages assumed. For example, one of our customers saw total scope 3 rise from 41,496 to 65,734 tonnes CO₂e after improving data accuracy in Category 1. But this isn’t necessarily a bad thing – improving the methodology is a critical step in delivering more accurate, reliable emissions reporting. 

Double counting when switching methods. When you add activity or supplier-specific data for a category, the corresponding spend must be excluded from the spend file. Failing to do this results in the same emissions being counted twice, once through spend and once through the more accurate data source.

Year-on-year comparability. Methodology changes make year-on-year comparison challenging. Under CSRD and SBTi, a change causing more than 5% variance in reported emissions may require a restatement of the baseline year. Document your methodology and any changes to it clearly each reporting cycle.

As one of our Climate Strategy Advisors puts it: “The first time when doing full three-scope coverage, I always see certain data gaps. Getting started with calculations, processing the data uploaded so far to assess preliminary results, that is really important.”