The shift happening in ESG reporting in 2026 is not primarily about which framework a company uses or how comprehensive its disclosures are. It is about whether the underlying data infrastructure can support what is being claimed. That is a different question from the ones sustainability teams have historically spent the most time on, and it is the one that is driving the widest gap between what companies publish and what investors and assurance providers can actually rely on.
ESG-labeled funds have exceeded $4 trillion globally. As that capital concentration has grown, so has investor scrutiny of the data behind the labels. The credibility problem is not just about greenwashing in the aggressive sense. It is about data that is inconsistent, estimated rather than measured, not traceable to primary sources, or produced by processes that have never been subject to any form of external review. A company can publish a thorough and well-intentioned sustainability report and still have data that would not survive a basic assurance engagement. The gap between those two things is where investor confidence is eroding.
What Assurance Actually Requires Versus What Most Sustainability Reports Provide
Limited assurance on ESG disclosures, the lighter of the two assurance standards, requires that the assurance provider conduct procedures sufficient to express a conclusion that nothing has come to their attention suggesting the disclosures are materially misstated. Reasonable assurance, the standard applied to financial statements, requires positive evidence that the disclosures are accurate. Under the EU's Corporate Sustainability Reporting Directive (CSRD), limited assurance is already required for companies in scope. The CSRD framework calls for escalation to reasonable assurance by 2028.
The preparation demands for even limited assurance are more significant than most companies have anticipated. Assurance providers need to trace disclosed figures back to primary data sources, understand the methodology used to calculate them, and assess whether the controls around that data collection process are adequate to prevent material error. For emissions data, that means documented measurement protocols, clearly defined organizational and operational boundaries, and evidence that the calculation methodology was applied consistently. For energy data, it means meter-level records that can be reconciled to reported totals. For water and waste data, it means source documentation that did not pass through a spreadsheet built by someone who no longer works at the company.
The operational reality for most corporate sustainability programs is that the data behind their disclosures lives in fragmented systems across business units, facilities, and geographies. Some of it is measured. Some is estimated using emission factors applied to spend or activity data. Some arrived from suppliers who produced it using methods that were not disclosed. The report synthesizes all of that into a single set of numbers. The assurance provider's job is to determine whether those numbers are reliable. The data infrastructure behind most current ESG reports is not built for that evaluation.
The Investor Side of the Problem Is About Comparability, Not Just Accuracy
The investor challenge with ESG data in 2026 is not only accuracy. It is comparability. A study cited by PwC found that a majority of investors believe ESG disclosures should be assured at the same level as financial statements. What that preference reflects is an expectation that ESG data should function like financial data: that two companies in the same sector, reporting on the same metric under the same framework, should be producing numbers that can be meaningfully compared by an analyst or a capital allocator.
That expectation is not being met. According to an ESG statistics survey published in April 2026, 47% of investors cite ESG data coverage gaps as their biggest challenge in integrating sustainability data into investment decisions. Coverage gaps is a polite term for data that is missing, inconsistent across reporting periods, estimated using methods that differ by company and year, or disclosed in ways that cannot be reconciled across companies even when they are using nominally the same framework.
The CDP 2026 disclosure cycle opened in June, with the scoring submission deadline in September. CDP remains one of the most widely used environmental disclosure systems for investors, buyers, and regulators. The 2026 cycle continues the platform's shift toward broader environmental integration, with expanded requirements across climate, forests, water security, biodiversity, plastics, and ocean-related impacts. Companies that have treated CDP disclosure as an annual data compilation exercise rather than a data quality management process are going to find the 2026 cycle more demanding than prior ones.
Where the Data Infrastructure Gap Is Widest
The data quality problems most likely to surface during assurance engagements cluster in predictable places. Scope 2 market-based emissions calculations that rely on energy attribute certificates or renewable energy contracts without verified additionality documentation. Scope 3 Category 1 purchased goods and services emissions estimated using spend-based methods because supplier-specific data is not available. Water withdrawal and consumption figures compiled from utility bills without reconciliation to operational records. Waste generation and diversion data provided by waste haulers using volume rather than weight measurements that cannot be converted reliably.
None of those are exotic data problems. They are the standard data limitations of most current corporate ESG programs. They have been adequate for stakeholder reporting because the audience was reading for narrative, not auditing for accuracy. That audience has changed. The investors, regulators, and assurance providers now engaging with ESG disclosures are reading with audit-level professional skepticism, which is a fundamentally different kind of reading.
The Measurement Question Is Going to Define the Reporting Conversation in Q3 and Q4
Sustainability teams heading into the second half of 2026 are going to face a version of the same question from multiple directions: can you demonstrate that this number is accurate, and how? That question will come from assurance providers preparing for first or second engagements. It will come from investors conducting due diligence on sustainability claims. It will come from buyers requiring Scope 3 data from suppliers. And it will come from regulators in jurisdictions where mandatory disclosure frameworks are moving into enforcement phases.
The companies that have invested in data infrastructure, documented methodology, and internal controls around ESG data collection are in a materially different position for those conversations than those that have invested in disclosure quality without the underlying data quality to support it. A well-written sustainability report based on poorly controlled data is not going to survive the assurance era intact. The gap between the two is where the next set of ESG credibility problems is forming, and it is forming now.
