Spend-based vs activity-based method in scope 3: when to use each

The spend-based method calculates emissions for a scope 3 category by multiplying spend in currency by a sector-level monetary emission factor, while the activity-based method multiplies a real physical quantity, such as kilograms purchased or kilometers traveled, by an emission factor specific to that activity. The first lets you start an inventory without depending on supplier data; the second is what the GHG Protocol requires to measure real reductions, and what verifiers and climate target standards end up asking for once a category turns out to be material.

What each method measures, per the GHG Protocol

The GHG Protocol's Corporate Value Chain (Scope 3) Standard and its Scope 3 Calculation Guidance describe both methods as valid ways to estimate emissions when there is no direct emissions figure from the supplier, but they are not interchangeable in terms of precision or in what they let you demonstrate.

Spend-based method

This takes the amount spent in a purchasing category (office supplies, consulting services, contracted transport, and so on) and multiplies it by an emission factor expressed in kg of CO2 equivalent per unit of currency, usually derived from sector-level input-output tables. It needs no supplier data at all: purchasing accounts classified by category is enough.

Activity-based method

This takes a physical quantity of the activity, such as tonnes of raw material purchased, kilometers driven by a carrier, kWh consumed or units manufactured, and multiplies it by an emission factor for that specific activity. Within this method there is a third, more granular tier: supplier-specific data, which uses the emissions intensity the supplier itself reports for its product or service, and is the most precise of the three.

When spend is an acceptable starting point

Spend is a reasonable starting point in a few situations:

• A company's first scope 3 inventory, when no supplier relationship exists yet through which to request activity data.
• Categories that, after an initial screening, turn out not to be material against the total footprint, where investing in better data is not yet a priority.
• Very fragmented purchasing spread across hundreds of small suppliers, where requesting activity data from each one is not operationally viable in a first cycle.

In these cases, the sensible move is to use spend-based explicitly and document it, not to treat it as a permanent solution.

Why it fails to measure reductions

The problem with spend-based accounting is not that it is imprecise at a single point in time: it is that it moves for reasons that have nothing to do with actual emissions. Two effects explain why:

• Price inflation. If the price of a raw material rises 15% from one year to the next without a single gram of the purchased quantity changing, the spend-based inventory will show higher emissions even though the physical activity is identical.
• Price changes from efficiency or supplier switching. Negotiating a better price, switching suppliers, or buying a cheaper but equally carbon-intensive product lowers spend, and with it the reported emissions figure, without any real reduction taking place.

That is why a company that sets a scope 3 reduction target and keeps measuring its main categories by spend can see a trend that reflects no operational change at all, in either direction. It is also why neither the SBTi nor an auditor will accept the spend-based method as a sustained basis for demonstrating a reduction in material categories.

How to move to activity data, category by category

The migration does not need to happen across all fifteen scope 3 categories at once: prioritize wherever the inventory carries the most weight.

Category	Spend data (tier 1)	Activity data (tier 2)	Supplier data (tier 3)
Purchased goods and services	Spend by accounting line item	Tonnes or units by material type	Supplier's product carbon footprint (EPD, LCA)
Transportation and distribution	Spend on contracted freight	Tonne-kilometers by transport mode	Actual emissions reported by the carrier
Use of sold products	Spend or units billed	Estimated lifespan and energy consumption per unit	Actual consumption measured or declared by the customer

The practical step from one tier to the next usually starts by asking the suppliers that carry the most weight in the category, not all of them at once, to provide physical quantities or, better still, their own footprint per unit of product. That is exactly the process we cover in the guide on how to collect scope 3 data from suppliers.

What verifiers and standards expect on data quality

ESRS E1 (the CSRD's climate standard) requires describing the methodology used to estimate scope 3 emissions, including the share of primary versus estimated data, and does not accept spend-based accounting as a permanent method for material categories without an explanation of the improvement plan. The SBTi's Corporate Net-Zero Standard requires companies with scope 3 targets to document a credible plan for improving data quality toward activity-based or supplier-specific methods, and does not allow excluding a material category simply because the data is hard to obtain.

In practice, an auditor reviewing a spend-based line item asks three things: which emission factor was used and where it comes from, whether that factor genuinely matches the good or service purchased rather than a loosely related generic category, and whether there is a documented plan to replace it with activity data in future reporting cycles. Without those three answers, the line item gets flagged as weak, even if the arithmetic is correct.

A simple numerical example

Imagine a company that buys 10,000 kg of a metal component for €50,000.

• Spend-based: €50,000 × a sector factor (say, a generic metals factor from an input-output table) gives an emissions figure tied to total spend, without distinguishing whether the metal is recycled steel or primary aluminum.
• Activity-based: 10,000 kg × a material-specific emission factor (steel versus aluminum, for instance) gives a figure that does reflect the material choice, and that does not move the following year if the metal's price rises 20% without a single extra kilogram purchased.

Neither of these emission factors should ever be taken from memory: each one you use needs to come from a recognized source (national input-output databases for spend-based, activity-factor databases such as those tied to the GHG Protocol or national inventories for activity-based), and the source needs to be cited in the inventory's documentation.

How to prioritize which category to migrate first

Before deciding which suppliers to ask for activity data, it helps to run a quick screen using the first year's own spend-based inventory: rank the scope 3 categories from largest to smallest in tonnes of CO2 equivalent, and start with the three or four that account for most of the total. Migrating a category that is not material to activity data consumes procurement team time without moving the inventory's result; migrating the heaviest category, even just for its top ten or twenty suppliers, does move it. This materiality screen, documented, is also the first thing an auditor asks for when reviewing why some categories were calculated in more detail than others.

Document the migration plan as such: which categories still run on spend this year, which are already on activity data, and the expected timeline for the rest. The migration does not need to finish in a single year, but you do need to be able to show the plan when asked why a material category is still calculated by spend.

Frequently asked questions

Can I mix spend-based and activity-based within the same category?

Yes, and it is common practice: use activity data for your largest or most material suppliers, and spend for the long tail of smaller suppliers, as long as you document what share of the total was calculated with each method.

Can the spend-based method be used to set a reduction target?

Not on a sustained basis for material categories: because it depends on price rather than physical activity, it cannot demonstrate whether a reported reduction is real. It works as a starting baseline, not as the method for tracking the target.

Which is more expensive to implement, activity data or supplier data?

Activity data is usually faster to get because it often already exists in procurement or logistics systems; supplier-specific data requires the supplier to have its own inventory or product footprint calculated, which takes longer but gives the most precise figure.

How often should the emission factors in use be reviewed?

At minimum whenever the source databases release a new version, and always before each reporting cycle, to avoid mixing factors from different years within the same inventory.

What if I have neither itemized spend nor activity data?

That is a more basic data-quality problem than the choice of method: it is worth solving first, following the process in our guide on how to calculate a carbon footprint with incomplete data.

Automating this move from spend to activity data, with the source of every factor documented, is one of the core functions of our carbon footprint tool.