ESG Trends 2026: What Changed in 2025 and What EU Companies Face Next

ESG reporting in Europe did not get bigger in 2025. It got narrower, later and more contested. The EU Omnibus package rewrote who has to report and when, the CS3D due diligence regime was cut back, and the technology debate moved from ambition to a plainer question: can the data actually be extracted from a filed report? This page is a year pivot for EU filers: what changed in 2025, what more than 1,400 European sustainability reports show when read as data rather than as prose, and what 2026 requires. If you came for the metric definitions themselves, my list of seven key ESG metrics covers that ground in full.

Last reviewed: 4 September 2026.

What Changed in 2025: Omnibus, Deferral and Divergence

The single most consequential development for EU filers was the Omnibus revision of the CSRD. Reporting duties now apply to companies above both thresholds, more than 1,000 employees and more than 450 million euro net turnover, for financial years starting in 2027, with first reports due in 2028. The staggered wave logic that dominated planning for three years no longer describes reality: what used to be CSRD wave 2 has largely dissolved into a single threshold test. CSDDD was narrowed in parallel to groups above 5,000 employees and 1.5 billion euro turnover, and its Article 22 obligation to put a transition plan into effect was deleted.

What More Than 1,400 European Sustainability Reports Actually Disclose

Most trend pieces quote what companies announced. I read what they filed. Across the European CSRD and ESRS reports in my extraction pool, covering the 2024 and 2025 reporting years, 69 percent carry all three emission scopes in machine readable form. A further 13 percent contain no extractable scope data at all, despite a filed sustainability report.

Looking only at the 1,401 reports from the 2024 and 2025 cycles, the quality picture gets sharper. About 8 percent of reports that carry a Scope 3 figure show it as smaller than Scope 1 or Scope 2, which is a methodological red flag, since Scope 3 is typically five to ten times larger.

My position after reading that pool: compliance without a time series is a photograph, not a steering instrument. A report that cannot be compared with last year cannot inform a decarbonisation decision, and an auditor will find that out faster than a stakeholder will.

One example from practice. An industrial manufacturer with roughly 900 employees in southern Germany published a complete looking report whose Scope 3 covered purchased goods and business travel only. The number was not wrong, it was incomplete in a way no reader could see, because the boundary was never stated.

Technology and Data: What Actually Held Up in Practice

Of the technologies that dominated ESG conference agendas, one category survived contact with a reporting deadline: unglamorous data plumbing. Extraction from ERP and utility systems, emission factor management with versioning, and a documented audit trail from source record to disclosed figure. These changed how data is collected. Ledger based ESG verification and sensor dashboards mostly did not, because the binding constraint was never trust in a number, it was that the number did not exist yet.

Scope 3 and value chain data remain the bottleneck. Spend based estimates get a first figure on the page quickly and then stop improving, because the factor, not the activity, drives the result. Moving the largest categories to activity data is the only step that makes a year on year series meaningful.

A point from building reporting software rather than from vendor briefings: market based Scope 2 accounting is methodologically correct and still fails in tool practice more often than any other method, because it depends on contractual instruments that companies cannot evidence per site and per year. If the certificates are not documented, the location based figure is the honest one.

A logistics group with about 400 employees in Austria spent one reporting cycle on a sensor pilot and the next on supplier master data. Only the second one changed the report.

What to Prepare for in 2026: Wave 2, VSME and Transition Plans

2026 is the year to build the dataset that the 2027 financial year will consume. A short checklist:

• Re-run the scope test against the Omnibus thresholds, subsidiaries included, and document the result.
• Fix baselines and reporting boundaries now, and give every datapoint one named owner.
• Decide between VSME and full ESRS for entities that fall out of direct scope but stay in someone else's value chain.
• Put the transition plan on a financing footing. Even without the Article 22 duty, lenders price it.
• Check every net zero and climate neutral claim against the Empowering Consumers Directive, which applies from 27 September 2026 and turns greenwashing into a litigation risk.
• Assess whether biodiversity and natural capital are material, and if so start collecting a year early.

Investors have shifted their asks in the same direction. Under stewardship code expectations they now want a comparable series, a stated boundary and a transition plan that reconciles with capital expenditure, rather than a broader set of indicators.

EU and US Divergence: Where the SEC Rule Leaves EU Filers

The SEC climate disclosure rule is not in force, and the Commission stepped back from defending it in litigation, so there is no federal US reporting duty that mirrors ESRS. For EU subsidiaries of US parents this creates a specific problem rather than a saving. The group has no consolidated climate dataset to hand down, while the European entity still needs one, either directly or because a customer asks for it under the value chain cap.

A German subsidiary of a US industrial group, around 600 employees, built its own scope inventory in 2025 for exactly that reason. The group finance system had nothing to give it. If your parent is American, plan the data work locally and do not wait for it to arrive from headquarters.

Frequently Asked Questions