Avoided emissions vs reductions: what Scope 4 can and cannot claim

A product team wants to publicize that its new offering helps customers cut emissions. The sustainability lead has to answer a harder question: does that benefit belong anywhere in the company's own carbon inventory? The short answer is no, and the confusion around that answer is exactly where avoided-emissions claims get companies into trouble. Avoided emissions, sometimes called Scope 4, are a real and useful concept, but they follow a different accounting system from your reductions.

Getting this boundary right matters because the two are constantly conflated. An avoided-emissions figure looks impressive next to an inventory, and mixing them, even innocently, is one of the fastest routes to a greenwashing accusation.

What avoided emissions are

Avoided emissions measure the emissions impact of a product relative to a world where that product does not exist. The World Resources Institute framework that popularized the concept defines it as the GHG impact of a good or service relative to the situation where that product does not exist, with positive differences frequently called avoided emissions. A more efficient motor, a low-carbon material, or a software tool that cuts a customer's energy use can all produce avoided emissions in the customer's operations.

The label Scope 4 is informal. It is not a formal GHG Protocol scope, and the WRI paper itself is framed around comparative emissions impacts rather than a new scope number. It is fine to use the term as shorthand, as long as everyone understands it is not part of the standard inventory.

The rule: never subtracted from your inventory

This is the point most teams get wrong. Avoided emissions must be reported strictly separately from a company's Scope 1, 2, and 3 inventory, never netted against it. The WBCSD's guidance is explicit that companies should not present avoided emissions as offsets to their inventory emissions or use them as part of net-zero claims, and should keep them separate from offsetting and carbon credits.

The Science Based Targets initiative draws the same line. Avoided emissions fall under a separate accounting system from corporate inventories and do not play a role in target-setting or achievement with the SBTi. Your target has to be met through reductions inside your own boundaries or value chain, not by pointing to emissions your customers avoided.

Where the credibility risk lives

Because the whole figure depends on a counterfactual, the baseline scenario is where credibility is won or lost. The WRI framework measures impact against a defined reference case, and it distinguishes attributional from consequential methods. When WRI reviewed claims from more than 300 companies, it found widespread methodological problems: none of the companies explicitly considered how their products might also increase emissions, and most claims lacked the information needed to assess their credibility.

The trend is large enough to matter. WRI noted that 36% of companies in a 2017 CDP survey claimed to sell products that help customers avoid emissions, up 20% from the year before. An inflated or self-serving baseline turns that claim into a liability. The WBCSD's 2023 guidance, updated to a version 2.0 in July 2025, is built around credible, transparent, and conservative assessment precisely to reduce that greenwashing risk, and sets eligibility gates requiring a real, science-aligned, and significant decarbonizing impact against a documented reference scenario.

Additionality and the counterfactual

The eligibility gates work as an additionality test. To claim that your product avoided emissions, you have to show that the alternative scenario, the world without it, would genuinely have produced more emissions, and that your solution is the reason for the difference. The reference scenario has to rest on documented, defensible assumptions rather than the least efficient product imaginable. Choosing an unrealistic baseline is the single most common way an avoided-emissions figure becomes indefensible, because a reviewer can simply substitute a more reasonable counterfactual and watch the claim shrink.

Honest communication is the other half. Avoided emissions should be reported with the boundary, the reference scenario, and the assumptions visible, and never headlined in a way that implies your own footprint is smaller. The safest framing states the avoided figure separately, alongside the actual inventory, so a reader can see both the contribution your products make and the emissions you are still responsible for.

Avoided emissions, offsets, and reductions are three different things

It helps to keep three mechanisms distinct. A reduction is a decrease in emissions within your own inventory boundary. An offset or credit is a reduction achieved elsewhere that you buy to compensate for your own emissions; under GHG Protocol project accounting, internal reductions need not be reported separately unless they are sold, traded, or used as an offset or credit. Avoided emissions are neither: they are the benefit your product creates in someone else's operations, reported on their own.

For the offset side of that table, our explainer on RECs, offsets, and real reductions goes deeper on how compensation claims work and where they mislead.

How clean baseline data supports real reduction claims

The distinction sounds academic until you try to prove a reduction. A reduction is a change against a baseline, and a baseline is only as good as the consumption data behind it. The International Performance Measurement and Verification Protocol (IPMVP) exists for this: because energy savings represent the absence of consumption and cannot be measured directly, they are derived by comparing metered energy before and after a change and adjusting for conditions such as weather and output.

That method depends on a clean, complete, normalized consumption history. If your baseline data has gaps, unit errors, or unadjusted weather effects, your reduction claim is as shaky as an inflated avoided-emissions baseline. The symmetry is worth noticing: whether you are estimating avoided emissions against a counterfactual or measuring a reduction against a historical baseline, the credibility of the claim collapses to the quality of the reference data. One is a hypothetical scenario, the other is your own past consumption, but both have to be documented, defensible, and free of the gaps that let a skeptic rework the number. For the mechanics of doing this rigorously, see our guide to measurement and verification with IPMVP.

Used honestly, avoided emissions still have a place. They help steer research and product decisions and show how a solution portfolio contributes to decarbonization, distinct from target-setting. Mission Innovation, for instance, showcased 100 innovations with the combined potential to avoid almost 3 gigatonnes of CO2e per year by 2030. The discipline is to report that number where it belongs, separately, and to build your inventory reductions on data solid enough to defend.