What I Found When I Actually Looked at ESG Data Up Close

The Gap Between the Report and the Reality

There is a version of ESG investing that sounds almost frictionless. Companies publish sustainability reports. Rating agencies read them, apply their methodology, and produce a score. Investors use that score to allocate capital. Outcomes improve.

Spending significant time with ESG data for my MSc research changed how I think about that picture. The data is real and the directional signals are meaningful, but the process between disclosure and investment decision is considerably messier than most of the marketing around ESG products would suggest.

Where ESG Data Comes From

ESG data comes from two main sources: what companies choose to disclose, and what third-party agencies infer or calculate from that disclosure and other public information.

The disclosure side is the foundation, and it has serious limitations. Until recently, there was no single mandatory global standard that applied to all publicly listed companies. The IFRS sustainability standards, S1 and S2, introduced in 2023, are moving things in a better direction, but adoption is gradual and not universal. Before those frameworks, companies could largely choose what to report, which framework to report against, and how much detail to include.

TCFD, GRI, CDP, SASB: each sets different requirements, uses different definitions, and focuses on different aspects of sustainability. A company reporting against GRI looks different from the same company reporting against TCFD, even if the underlying reality is identical. This is not dishonesty on the part of companies. It is a structural problem with voluntary and fragmented disclosure systems.

The Rating Disagreement Problem

For my FTSE 100 research, I worked with data from a single rating provider and held that constant throughout. I did this partly because it is the only honest approach for comparative analysis, and partly because I had seen how much disagreement exists between major providers.

Research on rating agency divergence has found correlations of around 0.6 between major ESG rating agencies, sometimes lower. That means roughly 40 per cent of the variation in how the same company is rated is unexplained by any shared methodology. MSCI and Sustainalytics can give the same company substantially different scores, not because one is wrong, but because they have made different choices about which factors matter most and how to weight them.

An agency focused on financial risk management weights governance heavily. One focused on impact weights environmental metrics differently. These are legitimate methodological choices, and they lead to very different outputs. Anyone using ESG ratings as if they were objective measurements of sustainability is working with a more uncertain input than they may realise.

The Disclosure Quality Problem

Even within a single reporting framework, the quality of disclosure varies considerably across the market. Large FTSE 100 companies have dedicated sustainability teams, external advisors, and board-level attention to their ESG reporting. Smaller listed companies often do not.

The result is that ESG data is systematically stronger and more reliable for larger companies than for smaller ones. This creates a bias in any analysis that does not correct for company size. The companies that are easiest to rate are not necessarily the most sustainable. They are the ones with the most resources to invest in reporting.

Why It Still Matters

None of this means ESG analysis is not worth doing. It means it needs to be done with an honest understanding of what the data can and cannot tell you.

The directional signal is real. Companies with strong governance practices, transparent environmental disclosure, and genuine stakeholder accountability tend to perform differently over time than those without. The challenge is extracting that signal cleanly from data that is incomplete, inconsistent, and sometimes shaped more by disclosure resources than by underlying performance.

This is why the standardisation conversation matters. The ISSB frameworks are a genuine step forward. When more companies are reporting against shared standards with shared definitions, the data will improve, and the analysis built on top of it will become more reliable.

For now, if you are using ESG data in financial analysis, the most important question to ask is: what is this score actually measuring, and what are its limitations? The answer will vary by provider, by sector, and by how recently the data was updated. That is not a reason to stop using it. It is a reason to use it carefully.