ESG data: What part can they play?

ESG data requirements grow sharply

As sustainable transformation gathers pace, ESG (Environmental, Social and Governance) data is becoming a pivotal factor in business life, helping to assess companies’ overall performance and long-term viability.

As with legal and financial information before it, ESG information provides an objective basis for dialogue between companies and their stakeholders. It enables investors and creditors to assess their risks and opportunities in the face of energy, ecological and societal transitions.

The demand for ESG data is set to increase rapidly over the next few years, for at least two reasons:

1. First, compliance with new regulations.
2. Second, the urgent need for business models to reflect these changes, as ESG data will become more and more of a sovereignty and operational continuity issue, both for companies (identifying priority areas for transformation, allocating resources efficiently, managing transition plans, etc.) and for investors (directing capital towards activities that are best aligned with the Paris Agreement, and that are most sustainable and profitable over the long term).

ESG data will evolve into a more advanced market, with suppliers and measurement methodologies becoming more and more specialized, in response to the complexity, multiplicity and interdependence of the issues at stake (human, climatic, socio-economic, etc.).

Some players will rely on imperfect, but rapidly and massively accessible data (ESG databases available “off the shelf” by sector, geography…), resorting to proxies and estimation models – and risking disconnection from the field. Others may prefer to use data as close to reality as possible (for example, using sensors or other tools).

Sustainable transformation will increasingly rely on vast chains of ESG data. Collecting, checking, processing and securing this data will involve costly resources, creating and accentuating inequalities between large and small companies.

Information systems (IS) will need to adapt accordingly. Players whose digital transformation is already underway will enjoy a competitive advantage in this new landscape.

The future roles of the different players remain to be invented, between EPM (Enterprise Performance Management) tools, ESG data platforms and specialized systems such as CMS (Carbon Management Systems).

Some companies are considering creating ESG “data lakes” (centralized repositories for structured and unstructured data at various scales), and are starting to work on linking their procurement and supplier information systems, to secure ESG information reporting on Scope 3 (greenhouse gas emissions indirectly linked to a product/service, upstream and downstream of the life cycle). Future concentrations will allow each link in the value chain (producers, auditors, ESG data analysts, etc.) to benefit from economies of scale.

Trust frameworks are still a work in progress

Beyond demand’s quantitative growth, investors, funders and insurers are setting ever higher standards for ESG data quality.

Their main focus? Understanding and comparing each company’s specific performance according to a range of criteria (business sector, size, geographical location, etc.).

In today’s competitive environment, reliable and comprehensible ESG data is a powerful asset in securing funding and capital. Markets will require higher levels of auditing and homogeneity, to improve fair comparison. Note: the degree of comparability differs from one topic to another (e.g. while standardization is well advanced for carbon, it is far more complex in the case of biodiversity).

Establishing trustworthy frameworks requires internal processes to be clarified, and independent auditors, agencies and assessors to be involved. The latter act as trusted third parties to ensure data quality and consistency, as well as the integrity of calculation methods.

Large auditing firms will play a necessary, but insufficient role. A network of independent, smaller players is essential to the resilience of the system as a whole – for example, to assess data produced by SMEs, which account for most of the economic tissue and supply chains.

Some analysts are predicting a growing number of legal disputes and an increasing role for ESG data in the courts, both from NGOs and investors concerned about being misled (for example, by an underestimation of the CAPEX level required to achieve a carbon-neutral target).

Interpretation skills will make the difference

At the crossroads between quantity and quality of ESG data, its analysis and interpretation remain an open question.

A significant part of ESG data’s value lies in its ability to be interpreted in a specific context (industry, geography…), considering margins of error that should clearly be displayed (and could be systematically communicated in the future, becoming ESG “meta-data”).

Rating agencies and investors will adapt their ways of working, enlisting increasingly specialized analysts capable of providing perspective on companies’ ESG data, moving on from quantitative scoring to more forward-looking and predictive insights. Activities involving the collection, verification, comparison and audit of ESG data are set to accelerate their trend towards specialization.

In the early 2000s, almost all ESG reporting was qualitative. Today, in a world more complex to read, we need to add quantity, which may (wrongly) give the impression that the data is easier to interpret. It’s a daunting challenge… all the more so as it is the exact opposite of financial data, which started with unified quantitative data and then added qualitative analyses.

Technology (artificial intelligence, blockchain, etc.) will play an ever-increasing role in producing, comparing, analyzing and adding value to ESG data, ensuring its traceability, but also making it available faster and more easily. Tools such as ChatGPT or Copilot will provide part of the answer, even if use cases have yet to be invented. Machine learning, meanwhile, can link a large number of parameters and information to establish the reliability of ESG data in relation to a company’s specific situation.

However, AI raises 5 main issues related to the assessment of companies’ ESG performance :

1. Reliability of primary datasets
2. Models’ transparency
3. Adaptability to sector-specific needs
4. Risk management (in relation to decision-making processes)
5. Human feedback (to ensure a balanced assessment and limit bias)