The Next Generation of Environmental Data—Built with Users

As a non-profit association, all of our license revenue is reinvested into the quality and completeness of our data, and into the innovation needed to keep meeting user needs. Our database curation process is already best in class, but we want to raise the bar further.

We know that to keep meeting user needs, we need to truly understand what those needs are, not run on assumptions. That’s why we’re pleased to share that we recently completed beta testing key components of the next generation of the ecoinvent database.

Throughout 2025 and 2026, we invited a diverse group of participants selected to represent the breadth of the ecoinvent user community. They spanned academic researchers, software developers, and enterprise sustainability teams. Together, they tested our methodology, challenged our assumptions, and told us what they needed.

Participants overwhelmingly confirmed our strategic direction, which includes:

1. Greater geographic precision: We will expand trade data integration and advance geographic modeling, enabling more accurate country-level supply chains. This means capturing not only where activities occur, but how local environmental conditions shape their real-world impacts.

2. Significantly increased scale: Our vision is to use advanced technology and AI to accelerate dataset creation across more sectors and regions without compromising the scientific rigor and transparency our users depend on.

3. Greater flexibility and adaptability: A restructured database architecture, standardized formats, and enhanced API access will make ecoinvent data easier to integrate, adapt, and apply—whether for compliance reporting, procurement decisions, or custom analytical workflows.

Our hypotheses were met with enthusiasm across the board.

Beta testers were particularly enthusiastic about increased regionalization and supply chain transparency. Users across all segments said the same thing: global averages are no longer sufficient. Their sustainability decisions, product carbon footprints, CSRD filings, and procurement optimization all depend on knowing where things are actually made, how they actually move, and what local conditions apply.

Being able to make things more closely represent the specific value chains that you want — that seems like a step forward in this version.

— James Joyce, Watershed

A second signal was about the process itself. Participants noted that being involved, having their feedback taken seriously, and seeing the changes they’d suggested appear in the next iteration was itself meaningful.

I really like that you opened this discussion to explore and collect suggestions about the new methods. I feel part of this because I’m going to use this data.

— Massimo Collotta, Henkel

ecoinvent’s Vision

We are building the fourth generation of the ecoinvent database to support evolving user needs and rapidly expanding use cases, and to lay the foundation for how sustainability data will be created and used for the next decade.

Features available on the release of 4.0 will include deeper regional granularity, more complete supply chains, and greater coverage across industries and geographies. It will include 100,000 datasets—four times what exists today. This increase is largely driven by granular trade modeling that was not possible in v3. This means we will know, for example, not just that a product was manufactured somewhere in Europe, but the specific country where it was produced, as well as the trade relationships that connect it to global supply chains.

Looking further ahead, v4 lays the foundation for a fundamental shift in how sustainability data is created and used. Data and calculations will be structured as models rather than static snapshots. This means it will appear less like an annual printed report and more like a live spreadsheet, where the underlying data can be updated, recalculated, and tailored to reflect real-world conditions as they change.

Version four moves the industry closer to dynamic, decision-ready insights that reflect each organization’s specific situation.

Why This Matters

Your sustainability reporting is under more scrutiny than ever. That’s where the quality of your underlying data becomes critical. As free alternatives multiply and AI-generated emissions factors enter workflows, the question is no longer whether you have data—it is whether the people reviewing your reports will trust it.

The beta testing process was an expression of something we believe: that the best way to build data infrastructure people can trust is to build it with the people who depend on it.

The strategy is clear, and I think it’s the right strategy for the future given the new expectations of the sector.

— Didier Beloin-Saint-Pierre, Empa