Looking beyond direct emissions: Why avoided emissions need better methodologies, and better collaboration

Explore

Insights

20 October, 2026 - Climate investing has come a long way in measuring environmental impact. Scope 1, 2 and 3 frameworks now give investors, businesses and regulators a common language for the emissions a company generates through its own operations and value chain. But one of the most important dimensions of climate impact still falls outside these measurements.

Many climate technologies create their biggest contribution not by lowering their own operational emissions, but by preventing emissions elsewhere. A heat pump replacing a gas boiler, a battery system enabling more renewable energy on the grid, an industrial process that uses less energy: all of these generate impact by avoiding emissions that would otherwise have occurred. These are called avoided emissions, sometimes referred to as Scope 4.

For a climate-focused investor like us, this matters. A young venture can have a modest operational footprint and still be one of the most decarbonising companies in a portfolio, precisely because of how much it helps others avoid. Looking only at operational emissions risks missing where a lot of the impact is happening.

The problem is that avoided emissions cannot be measured directly, the way Scope 1 to 3 emissions can. Every calculation rests on a question: what would have happened without this technology? Which alternative would customers have used instead, and what would that have cost the climate? The answer to that question drives the entire result, which makes avoided emissions as much a methodological challenge as a data one.

No shared standard

Interest in avoided emissions is growing fast. Investors want to understand the impact their portfolios enable. Ventures want a credible way to show the value they create. Yet, there is still no single accepted way to measure it.

We do not see that as a weakness. The field is still young, and that gives investors, ventures and industry bodies room to help work together towards setting good practices. That is also why we carried out our own research into the topic, led by Dante Wiersma. We combined a review of the avoided emissions methodologies currently used in private markets with interviews with eight portfolio ventures, funds, and other partners who work with avoided emissions reporting. Our goal was not to find one correct methodology, but to understand where current approaches agree, where they differ, and how reporting can become more transparent, comparable, and workable for early-stage ventures.

What we found: two strong frameworks, and six recurring problems

Two frameworks currently carry the most weight in private markets. The WBCSD Guidance on Avoided Emissions is the most comprehensive option available, with detailed, sector-specific reference scenarios. Project Frame is a newer framework built specifically with early-stage ventures in mind. In our interviews, people consistently preferred Project Frame because it takes the realities of early-stage companies seriously: limited data, limited resources, limited in-house expertise. Tellingly, nobody described either framework as complete on its own. In practice, every assessment ends up being a bespoke exercise that differs on a case-by-case basis.

Six challenges shaping avoided emissions reporting

Different organisations use different methodologies, but the challenges they run into are similar. Across the research, six themes kept emerging.

1. Determining who can credibly report avoided emissions

Many methodologies ask organisations to meet eligibility requirements before they even start reporting avoided emissions: setting climate targets, getting independent verification, or showing that the technology’s impact is material. These requirements exist for good reason. They filter out fake impact claims and give reported figures more credibility.

In practice, however, they create friction. Early-stage ventures often do not have the budget, in-house expertise or operational maturity to meet all three requirements, and holding a young climate-tech company to the same standards as large companies risks turning a useful filter into an unnecessary burden.

We think the fix is proportionality, not a lower bar. Requirements should scale with a venture’s materiality and capability: younger ventures should show that their technology creates meaningful impact, while more mature companies can be expected to set formal climate targets and get independent verification. That way, reporting quality can improve as a company grows, instead of discouraging it early on.

2. Establishing a credible baseline

If one methodological choice decides the outcome of an avoided emissions calculation, it is the baseline. Avoided emissions are calculated by measuring the impact of the new technology compared to the alternative scenario: what would most likely have happened if the technology had not existed?

Building that scenario is difficult and open to assumptions as to what is being replaced. Some organisations construct it through customer interviews, others lean on industry experts, tools such as CRANE, or existing market data. Many funds settle on one fixed baseline; others build several to reflect different market conditions and customer behaviour.

None of these approaches is inherently right or wrong. What matters is how transparently they are built and explained.

Rather than prescribing a single method, our research points to a consistent process: identify the impact pathway, check the baseline against multiple sources, explain the reasoning behind it, and revisit it as the technology and market evolve. In the end, being transparent about how a baseline was chosen is worth more than claiming certainty about the number itself.

3. Working with imperfect data

Data availability often came up as one of the biggest practical obstacles. Most early-stage ventures do not collect the primary operational data a full Life Cycle Assessment would need, and their customers usually do not measure it either. As a result, most organisations end up relying on secondary data: scientific literature, industry databases, or external consultants.

Resources such as EcoInvent and DEFRA are useful here, but they have limits too. The data does not always match the technology being assessed, geography can affect relevance, and access is often expensive or restricted. What came through clearly in the interviews is that imperfect data should not stop organisations from measuring impact. It should push them to be more transparent about the quality of what they are using and to introduce measurement protocols to start collecting internal data.

We would suggest a simple data hierarchy: verified primary data first, then reputable third-party databases, then peer-reviewed literature, and industry averages only as a last resort. Whichever level ends up being used, it should be stated openly alongside the figure.

4. Making Life Cycle Assessments comparable

LCAs sit at the centre of most avoided emissions calculations. They quantify a product’s environmental footprint across its lifecycle, which is what makes comparing a new technology with its conventional alternative possible. Our research found a recurring problem here too: making sure the scopes actually match.

An LCA can be made using different scopes: cradle-to-gate, cradle-to-grave or cradle-to-cradle. If the new technology’s LCA uses one scope while the baseline uses another, the comparison stops meaning much. This can happen when organisations reuse a published third-party LCA. Moreover, differences in geographies between LCAs can also reduce reliability.

This should not discourage the use of existing LCAs. It just means checking that the product and its baseline are assessed on comparable terms, and citing the sources and assumptions behind any LCA that is reused.

5. Attribution

The hardest challenge we came across, by some distance, is attribution.

Climate impact is rarely the work of one organisation alone. A venture builds the technology, investors provide the capital, suppliers contribute components, customers adopt the product, and policymakers sometimes create the conditions that make adoption possible. All of these actors contribute to the same avoided emissions outcome.

The question is how to divide that impact. There is no broadly accepted method yet. Some investors attribute a share of avoided emissions based on their equity ownership; others skip quantitative attribution altogether and instead describe how they contributed.

Both come with trade-offs. Assigning impact to stakeholders risks double-counting, with several stakeholders claiming credit for the same reduction. Avoiding attribution altogether risks understating the real role investors play in scaling climate solutions.

We think the more balanced path is contribution rather than ownership. Instead of claiming a specific share of avoided emissions, organisations can describe how their investment, expertise or commercial support helped enable the outcome, alongside a clear account of what they actually did to be additional to that impact. Attribution is a complex problem without a solution, but this approach is a more honest way to talk about impact in the meantime, without risking double-counting.

6. Transparency creates credibility

One of the clearest findings across both the literature and the interviews is that credibility depends less on the figure itself and more on the methodology behind it.

Right now, organisations often publish avoided emissions as a single headline number with little explanation of how it was calculated. Without visibility into the assumptions, data quality and methodological choices behind it, comparing one figure to another becomes difficult. Across the interviews, people consistently said that more transparency would do more for the field than chasing more precision.

A few practical ideas came out of this: a shared place to deliver documentation during information requests, more regular conversations between funds and portfolio companies about what reporting actually works for both sides, and focusing on the three to five metrics that matter most, rather than trying to report everything.

Uncertainty is part of avoided emissions and will not go away. The point is not to eliminate it, but to make it visible: explain the assumptions, the limitations and the evidence behind every number. That is what enables collaboration between stakeholders to improve.

A tiered framework, not a fixed standard

Based on this research, we built a three-tier framework with proposed guidelines for each tier.

We see the tiers as a roadmap rather than a checklist. A venture knows what moving up looks like, and we can use that progression to raise the bar gradually, in step with a company’s growth rather than ahead of it.

In practice, this means using the framework in our regular conversations with funds, both during onboarding and through the annual impact reporting cycle, to agree where a venture sits today and what the next step up would look like. Our more mature ventures can also help newer ones through webinars and shared learning, so the network improves together rather than each company solving hurdles alone.

The framework does not solve everything. Attribution in particular remains an open question that will need more research before a standard emerges. But the underlying principle holds regardless: transparency matters more than precision. The context behind a number is more valuable than the number itself.

Better conversations before better calculations

Our research shows that real improvement is already possible through more transparency, more consistent methodology, and closer collaboration between investors, ventures and other stakeholders.

The point of avoided emissions reporting is not to produce a perfectly precise number. It is to give investors a clear picture of how a figure was built, what it assumes, and where the uncertainty sits. And then improve these factors together: get better data, make better assumptions, etc. We believe that comes from ongoing dialogue, not from waiting for a perfect standard to arrive. Getting this right will not happen in one step. It is something we intend to keep building together with our ventures, one conversation at a time.