Scope 3 often accounts for 80–90% of a company’s total emissions, and it is the part you do not directly control. This guide covers the boundary rules, the 15 categories, how to choose a calculation method, where the data comes from, how reduction actually happens, and what disclosure frameworks now require.
Six sections, from boundary rules to disclosure
The three scopes, in plain terms
Before Scope 3, the boundary rules have to be clear. This is where most first inventories go wrong.
The GHG Protocol splits a company’s emissions into three scopes. The split is not about importance — it is about who controls the source.
For most companies outside heavy industry, Scope 3 is the large majority of the total — frequently 80–90%. That is why customers, investors and rating agencies keep asking for it, and why it cannot be treated as optional.
One decision to make first: the organisational boundary. Equity share, financial control and operational control give different numbers for the same group. Write the choice down, apply it consistently, and disclose it — auditors and rating agencies both look for this.
The 15 categories
Scope 3 is not one number. It is fifteen buckets, eight upstream and seven downstream, and they are wildly uneven in size.
Two practical points. First, categories that do not apply can simply be reported as not applicable, with the reason stated — completeness is about honest scoping, not about filling in every row. Second, resist the urge to refine everything: estimate all fifteen roughly, then invest effort only where the numbers are large.
Choosing a calculation method
Accuracy and effort trade off directly. The right answer changes as your programme matures.
A spend-based first pass has one large advantage: it is achievable with data finance already holds. Its weakness is that buying the same thing more cheaply appears as a reduction, which is exactly the wrong incentive. So treat spend-based numbers as a map for prioritisation, not as a performance measure.
Moving to supplier-specific data is where real reduction begins, because it is the only method in which a supplier’s genuine improvement shows up in your inventory. It is also the most work, so sequence it: top suppliers by estimated emissions first, not top suppliers by spend.
Emission factors and data collection
The number you publish is only as good as the factors behind it. Know which level you are on.
Data collection sits on three levels, and most companies are on all three at once:
- Level 1 — secondary data. Published average factors by industry or product category. Fast, comparable, and blunt.
- Level 2 — refined secondary data. Regional or process-specific factors, chosen to match your actual inputs more closely.
- Level 3 — primary data. Measured values from your own operations or from named suppliers, ideally with third-party verification.
Established factor sources include the GHG Protocol’s own guidance and calculation tools, national datasets published by governments, and commercial life-cycle inventory databases. Whichever you use, record the source, the version and the year for every factor. Assurance providers ask for exactly that, and a factor updated between reporting years is a common cause of apparent movement that has nothing to do with real emissions.
Recalculation policy: decide in advance when you will restate a base year — typically for structural change such as acquisitions and divestments, or for a methodology change above a set threshold. Publishing the policy is what makes your trend line credible.
Reducing Scope 3, not just reporting it
Scope 3 falls through decisions taken with suppliers and customers, so the levers are commercial rather than technical.
Start with hotspots. In practice a handful of categories — usually purchased goods, capital goods, and use of sold products — account for the bulk of the total. Levers by category look roughly like this:
- Purchased goods and services: supplier engagement, material substitution, design changes that reduce input volume, and adding emissions criteria to sourcing decisions.
- Capital goods: lower-carbon construction materials, longer asset life, and refurbishment instead of replacement.
- Use of sold products: product efficiency. For most manufacturers this is a product-development question, not a sustainability-team question.
- Logistics: mode shift, load factor, and carrier selection.
- Investments: portfolio-level targets and engagement with investees.
Two cautions. Switching to a supplier with better data is not a reduction; be able to separate real change from data change when you explain a movement. And avoid resolving Scope 3 with offsets — most standards expect reductions in the value chain first, with credits reserved for residual emissions.
Where Scope 3 has to be disclosed
Reporting requirements have converged fast, and Scope 3 is now inside mandatory frameworks rather than voluntary ones.
The IFRS Sustainability Disclosure Standards issued by the ISSB require disclosure of Scope 1, 2 and 3 greenhouse gas emissions, with relief provisions in the first reporting period. In the EU, the Corporate Sustainability Reporting Directive and the ESRS require value-chain information from companies in scope, and Japan’s SSBJ standards follow the ISSB structure. CDP’s questionnaire asks for Scope 3 category by category, and SBTi target validation covers value-chain emissions.
The practical consequence is that the same inventory now has to serve several audiences at once: a regulator, a rating agency, a customer procurement team and an assurance provider. Build it once, document it properly, and reuse it — rather than assembling a different number for each request.
The questions that come up in every first inventory
Do we have to calculate all 15 categories?
No. Categories that are not relevant to your business can be reported as not applicable, provided you state why. What matters is that the scoping decision is explicit and consistent between years. The usual failure is not an omitted category but an undocumented one.
Our suppliers will not give us data. What do we do?
Do not start by asking for primary data. Estimate the category with average factors, identify the suppliers that dominate the estimate, and approach only those — with a template, a method and a deadline you have already worked out. Requests that make the supplier design the answer tend to go unanswered.
Our Scope 3 number moves every year. Is that a problem?
Only if you cannot explain it. Set a base year, publish a recalculation policy, and separate three causes when reporting a change: real emissions change, activity change, and methodology or factor change. Movement that is explained is credible; movement that is not looks like weak control.
Can we use carbon credits to cover Scope 3?
Not as a substitute for reduction. Mainstream frameworks expect value-chain reductions first, with high-quality credits or removals applied to residual emissions and reported separately. Presenting offsets as a Scope 3 reduction is a common greenwashing finding.