CDP Scope 3 relevance: how to run the relevance assessment and mark a category not relevant in 2026

CDP Scope 3 relevance is the assessment of which of the fifteen Scope 3 categories materially apply to your business. CDP asks you to account for all fifteen, not calculate all fifteen: each one is marked relevant and calculated, relevant but not yet calculated, or not relevant with a stated reason. In the 2026 questionnaire that reason became a scored drop-down rather than free text, which changes what a defensible screening has to look like.

Last updated August 2026. Most teams treat the relevance question as paperwork and find out in December that it was not. An incomplete screening blocks a Leadership essential criterion, and that criterion is a stated pre-requisite for the verification criterion, so one thin answer takes out two. Below: how the relevance test actually works, what changed in 2026, what counts as a defensible not-relevant answer, and the categories companies wrongly exclude.

What does CDP mean by Scope 3 relevance?

Question 7.8 of the full corporate questionnaire lists all fifteen Scope 3 categories and asks you to classify each one. There are three answers. Relevant, calculated means you consider the category material and you are reporting a number for it. Relevant, not yet calculated means you accept it is material but do not have the data this cycle. Not relevant, with an explanation, means you are asserting that the category does not apply or is not material to your business.

The distinction people miss is that relevance is a judgment about your business, and calculation is a judgment about your data. A category can be relevant and uncalculated without penalty, as long as you say so. What CDP does not accept is silence, or a blanket exclusion with nothing behind it. Leaving categories blank is the fastest way to stall a response at Disclosure level regardless of how strong the rest of it is.

The GHG Protocol sets the underlying test. Its Corporate Value Chain (Scope 3) Standard asks companies to account for all categories and report those that are relevant, judged against size, influence, risk, stakeholder expectations, outsourcing and sector guidance. CDP's question is that test rendered as a table. The categories themselves, one by one, are set out in our reference on the 15 Scope 3 categories.

What changed for Scope 3 relevance in the CDP 2026 questionnaire?

Question 7.8 gained a new column, "Reason for this category not being relevant", and CDP scores it. At Disclosure level, Route B now requires a drop-down option to be selected in that column. At Awareness, any option other than "Other, please specify" is required. At Management, the scoring that previously rewarded your free-text description in the "Please explain" column was replaced by scoring on the new drop-down, though the free text is still checked for completion.

That reads like an interface tweak and behaves like a requirement change. In 2025 a well-written paragraph earned the points, so a category could be excluded on a plausible narrative. In 2026 the paragraph is checked for presence only, and the points come from picking a reason off a list. There is no partial credit for a thoughtful explanation of a weak exclusion. CDP's own rationale for the change says it aligns the column with the GHG Protocol requirements for relevant Scope 3 categories.

The practical effect is that you now have to be able to name why a category does not apply, in a category CDP recognizes, before you fill the table in. That is a screening exercise, and it has to happen before drafting rather than during it. The wider set of 2026 scoring updates, including the governance and verification changes that landed in the same cycle, is on CDP 2026 changes.

How do you run a Scope 3 relevance assessment?

A relevance assessment is a screening estimate, not a calculation. You are trying to establish the rough size of each category quickly enough that the answer costs days rather than months, and precisely enough that you can defend the classification. The standard approach is a spend-based screen across the whole company followed by a judgment pass.

1. Pull a full year of accounts payable, coded by account or supplier. This is the single most useful dataset for a relevance screen because it covers everything the company bought, which is what most Scope 3 categories are.
2. Map spend to categories. Most lines land in category 1. Capital purchases move to category 2 according to your own capitalization policy, not to any emissions logic. Freight lands in category 4 or 9, travel in 6, and so on.
3. Apply a screening emission factor per dollar. The EPA Supply Chain Greenhouse Gas Emission Factors dataset covers 1,016 US commodities on six-digit NAICS codes and is expressed in kilograms of CO2e per dollar, which makes it the natural fit for a spend screen. Use the with-margins variant against purchase prices, since a purchase price already contains wholesale and retail margins.
4. Rank the categories by estimated size. Anything above a few percent of your total is relevant, full stop. The argument only starts below that line.
5. Apply the qualitative tests to the small categories. Size is one of six GHG Protocol criteria. A small category can still be relevant because of risk exposure, stakeholder interest, sector guidance, or because you outsourced an activity that used to be in your Scope 1.
6. Record the reason for every exclusion as you go. This is the part that is usually reconstructed months later from memory. Write it while the number is in front of you.

The whole exercise is only as good as the ledger you start from, which is where it tends to break. Purchase data is organized for paying suppliers, not for measuring emissions, so a single supplier can appear under four spellings and a general ledger account called "professional services" can contain three unrelated categories. Getting a clean, consistent year of spend out of an accounting system is most of the work. The step-by-step method for turning that spend into numbers is in calculating Scope 3 emissions from spend, and the four GHG Protocol methods you graduate to afterwards are on how to calculate Scope 3 emissions.

What counts as a defensible not relevant answer?

Three kinds of reason hold up. The activity genuinely does not occur: a software company has no processing of sold products because nothing it sells gets processed. The activity occurs but is already counted elsewhere: emissions from a leased asset sitting inside your operational control boundary belong in Scope 1 and 2, not in category 8. Or the activity occurs and is immaterial on a screening estimate, which is the only reason that requires you to have done the screen.

Two kinds do not hold up, and both are common. "We do not have the data" is not a relevance statement, it is a calculation statement, and the questionnaire has a separate answer for exactly that case. And "not applicable to our sector" without a screen behind it is an assertion, not a reason, which is precisely the gap the 2026 drop-down closes.

Category 11 is the one that decides whether a footprint is honest. For a manufacturer of anything that burns fuel or draws power, use of sold products is frequently larger than every other category combined, and excluding it produces a total that is defensible line by line and wrong overall. The boundary questions around who counts what are covered in double counting Scope 3 emissions, and the materiality method in more depth in the Scope 3 materiality assessment.

Does CDP publish Scope 3 relevance by sector?

CDP does not publish a definitive per-sector relevance list that overrides your own judgment, but it does apply sector-specific scoring and expects the categories that dominate your sector to be calculated rather than screened away. The GHG Protocol's technical guidance goes further, setting out which categories are typically most significant for different business models, and that is the document to cite when you need external support for a classification.

The pattern is stable enough to plan around. Retailers and distributors are dominated by category 1 and category 4. Manufacturers of powered goods are dominated by category 11. Professional services firms are dominated by categories 1, 6 and 7. Financial institutions are dominated by category 15, which is why they are assessed on portfolio impact instead of ordinary Scope 3 disclosure.

What happens if your Scope 3 relevance screening is incomplete?

It caps your score, and not in the place you would expect. EC-CC15, the Scope 3 disclosure essential criterion, requires base year emissions for every category you have marked either relevant and calculated or not relevant and calculated, plus real calculations for at least one category including your sector-relevant one. Miss it and Leadership is closed, so the best available result is a B.

Then it compounds. EC-CC15 is a stated pre-requisite for EC-CC17, the verification criterion. So a company that has paid an assurance provider to verify 95% of its Scope 1 and Scope 2 emissions, which is expensive and takes months, gets no Leadership credit for it if the Scope 3 table has gaps. That is the most costly sequencing mistake in a CDP response, and it comes from treating the relevance question as a formality. The full criteria list, level by level, is on CDP essential criteria, and what each resulting letter means is on CDP score.

How to make the relevance answers auditable

A relevance decision is a judgment, and judgments get challenged. Whatever you record has to survive being read by someone who was not there: an assurance provider sampling your basis of preparation, a customer's sustainability team, or your own successor next cycle. Three things make that work.

Keep the screening estimate, not just the conclusion. A note saying category 5 was excluded is worth very little; a note saying category 5 screened at 0.3% of total estimated Scope 3 on FY2025 spend using EPA supply chain factors is defensible and reusable. Record the basis alongside it: which factor set, which version, which reporting year. And tie each figure back to the underlying records, because that trail from a total back to a source document is exactly what a verifier samples and the usual reason a spreadsheet inventory fails its first assurance engagement.

Where supplier-specific data is the goal rather than spend factors, a growing number of suppliers now publish product carbon footprints and environmental declarations on their own websites, and collecting those across a few hundred vendors is a data-gathering problem rather than an emissions one. Turning scattered published pages into clean structured data is usually faster than emailing procurement contacts one at a time. Our own approach works from the other end, classifying accounts payable and utility records into a scope-by-scope inventory with every figure tied to the document that produced it, so the relevance screen and the eventual calculation come from the same ledger rather than two separate exercises.

This article summarizes the CDP full corporate questionnaire and its 2026 scoring changes, together with the GHG Protocol Corporate Value Chain (Scope 3) Standard and its Technical Guidance, as they stood in August 2026. CDP describes its 2026 criteria as provisional. Relevance is a company-specific judgment and the examples above are illustrative rather than determinative: check the exact scored data points against CDP's own documents and your own screening before you submit. Nothing here is legal or accounting advice, and this article is not affiliated with or endorsed by CDP.