Corporate biodiversity: what the first ACT Biodiversity roadtest reveals

In January 2026, EVEA published — on behalf of ADEME and the French Biodiversity Office — the report of the first roadtest of the ACT Biodiversity methodology. Thirteen large French groups underwent a full assessment of their biodiversity strategy, for the 2024 reporting year. The results offer a rare, quantified snapshot of the biodiversity maturity of the French economy.

A methodological caveat first: at the time of this roadtest, the ACT Biodiversity methodology was still in a draft version. The figures below should therefore be read as a development-stage status report. The version 1 of the methodology, published in June 2026, has since corrected part of the points identified.

Key takeaways

• The roadtest assessed 13 large French groups (construction, agri-food, chemicals, energy), for the 2024 reporting year.
• Average performance score: 32/100 (narrative C, neutral trend); best score 59, lowest 14.5, observed theoretical maximum 65/100.
• Weakest module: the upstream value chain, at 19 %, described as a "critical blind spot". The most mature: policy engagement, at 54 %.
• "Nature-positive" trend after expert judgement: 2 companies positive, 8 neutral, 4 negative.
• Major finding: public CSRD data was not enough to carry out the assessments, even for companies already subject to the directive.

The sample: 13 companies, 4 sectors

The roadtest covered 13 companies from four sectors: construction, agri-food, chemicals and energy. Most assessments relied on private data shared under a non-disclosure agreement (NDA) — a point we return to below, as it has heavy implications for CSRD reporting. Each company received a three-dimensional score: performance (0 to 100), narrative (A to D) and trend (+ / = / –).

An average performance score of 32/100

The most telling result: the final average performance score stands at 32 (out of 100), with an average narrative grade of C and an average neutral trend (=). The spread is wide:

* By aggregating, indicator by indicator, the best mark obtained across all participants, the report sets a practical ceiling of 65/100 — proof that no company, even combining the best practices observed, reached excellence on every front.

The Achilles' heel: the upstream value chain (19 %)

The module-by-module analysis is even more instructive. On average, companies score best on policy engagement: 54 %, their most mature dimension. Conversely, the weakest module is unambiguous:

• Upstream activities: 19 % — the lowest average score of all modules.
• Business model: 24 %.
• Intangible investment: 25 %.

Yet upstream (module 4) and direct operations (module 2) together account for 43 % of the performance score — the heaviest weighting in the methodology. A score of 19 % on the upstream link therefore heavily penalises the overall result. The report calls this link a "critical blind spot" and stresses that, for all participants, "collecting data for Module 4 was very difficult". The underlying observation, expressed by the companies themselves: most "were previously focused on their direct impacts, especially on their sites, and did not realize the importance of upstream supply chain".

The narrative grade: a deficit in transformation and risk management

The narrative score rests on five maturity dimensions, scored from 0 to 4. The roadtest averages show where the strengths and weaknesses lie:

Two lessons. First, companies are relatively comfortable on the quality of their data (2.38) — but this figure must be qualified by the fact that most relied on private, unpublished data. Second, biodiversity risk management (1.69) and business-model transformation (1.95) remain the weak links: few companies have integrated biodiversity as a structuring strategic risk.

(Reading note: the report states "2" for the Consistency & Credibility dimension in the text, while the corresponding summary chart may display a slightly higher value; we retain the textual value.)

The "nature-positive" trend: 4 companies out of 13 going backwards

The trend score indicates whether the company is moving toward or away from a pathway compatible with a "nature-positive" economy. The process is instructive: the ACT tool, applied automatically, initially classified 12 companies as "neutral" (=) and only one as "negative" (–) — a result deemed too lenient by the evaluators, the tool's level of requirement being calibrated very high. After adjustment through the analysts' expert judgement, the final distribution is as follows:

• 2 companies with a positive (+) trend;
• 8 companies with a neutral (=) trend;
• 4 companies with a negative (–) trend.

In other words: for four groups out of thirteen, there is no tangible indication of improving biodiversity performance in the years ahead. Only two companies show robust signs of transforming their business model.

The major lesson: public CSRD data is not enough

This is probably the most structuring finding of the roadtest for sustainability and finance teams. The report is explicit: the assessments "were mainly based on private data", which means that "even for companies publishing a sustainability report in line with the CSRD, public data was not sufficient for the evaluation".

The reason given: a still-real lack of maturity on the ESRS E4 (Biodiversity) standard. The report adds a broader observation: "in general, companies are less mature on biodiversity than climate change". Many have a climate transition plan, few a "nature" transition plan — even though the latter is a prerequisite to start the assessment.

What these figures tell companies

The first ACT Biodiversity roadtest paints an uncompromising picture: low overall maturity (32/100), a major blind spot on the upstream value chain, and above all a shortfall of reliable, traceable, usable data — including at companies already subject to the CSRD. This data shortfall is no methodological detail: it is the tipping point. A biodiversity strategy is only as good as the measurement underpinning it. That is the issue we address in our article on the role of independent certification in biodiversity measurement.

From raw data to certified data. The roadtest shows it: the difficulty is not producing a figure, but producing a reliable, comparable and defensible one. Effinature certification, delivered by IRICE — a body accredited by Cofrac under no. 5-0655 to ISO/IEC 17065 (scope available at www.cofrac.fr) — provides that guarantee at field level. Talk to an IRICE expert.

Source: ACT Biodiversity roadtest report (EVEA for ADEME and the OFB, January 2026), draft version of the methodology.