Double Materiality: What the Second Year of CSRD Reveals
After a first year spent getting to grips with the CSRD, the second year raises a new question: have companies fundamentally revised their double materiality analysis, or have they simply rolled it over? The 2026 KPMG study, based on over 50 reports from CAC 40 and Next 20 companies, provides precise, data-driven, and sometimes contrasting insights.
Temps de lecture estimé : X min
Third episode of the CSRD Summer Workbook: after retracing the three regulatory overhauls of the last three years, and then studying what Scope 3 reveals about the value chain, this week we turn to double materiality and what its second year of implementation is already showing.
91% of the companies in the panel revised their double materiality analysis for their second reporting cycle. The figure is reassuring on the surface. It masks a more nuanced reality: 61% carried out a targeted revision, focusing on specific issues or IROs, and only 30% conducted a substantial revision, questioning the very structure of their matrix.
A company that carries over its previous year's matrix with minor adjustments checks the "update" box without truly questioning its relevance.
The auditor, however, is looking for something else: a record of the methodology used to decide what needed to change, and why.
External consultation, the blind spot
Of the companies that updated their DMA, only 3 out of 10 consulted external stakeholders as part of this process. The majority did not reopen the dialogue with the stakeholders who had informed the initial analysis—and the reports do not specify what the updates were based on in these cases.
This choice may be justifiable.
However, the absence of external consultation must be explained, not simply noted by its omission in a report that remains silent on the matter.
Who validates, and how can it be proven
Double materiality governance took shape in the second year, with a clear distribution of validation roles:
- Board of Directors (23%),
- General Management (16%),
- Audit Committee (15%),
- Executive Committee (15%),
- Audit and Risk Committee (11%),
- Others (11%),
- Sustainability Committee (8%).
This diversity shows that there is no single expected governance model. What the auditor does verify, however, is that the body validating the double materiality results is identifiable and that the process leading to this validation is documented.
Fewer IROs, not less materiality
The average number of IROs per company has fallen by 25%, from 52 in 2024 to 40 in 2025. This decrease reflects a streamlining effort following an initial year often characterized by overly long lists, rather than a decline in ambition.
However, it comes with a caveat: one in five companies still fails to clearly describe which material issue each of its IROs relates to. Reducing the number of IROs without clarifying this link undermines the very clarity that the streamlining was intended to provide.
What is needed to document double materiality in year 2
The formalized update method: what has been reviewed, what has remained unchanged, and on what basis
- An explicit justification in the absence of external consultation
- The governance body that validated the result, and the record of the process that led to it
- The link between each retained IRO and the material issue it relates to
What Harnest documents for you
These four points of vigilance correspond to what Harnest tracks natively within the tool.
Every sustainability issue maintains a history of changes made to its attributes and IROs, which can be viewed via the Audit icon—providing exactly the methodological proof that auditors look for.
Stakeholder engagement campaigns are recorded and linked to the relevant issues, meaning that a lack of external consultation becomes a documented choice rather than a blind spot.
In Harnest, corporate governance bodies are treated as distinct entities linked to the issues they approve, making the validation chain traceable from one reporting cycle to the next.