Empowering the Financial Analyst With ESG Through Analytics and Data Science

Dataiku Product Sophie Dionnet

The race to ESG continues to accelerate in the financial services space. All major players have now taken firm commitments to embed Environmental, Social, and Governance (ESG) criteria in all their critical processes, with a strong focus on global warming management. The creation of the Net-Zero Alliance by 43 major banks including Société Générale, Citi, and Morgan Stanley and main insurance players including Aviva and Zurich insurance is yet another sign that the financial industry is getting organized to play its role in taking up the climate change challenge. The impact financial players can have is also more and more visible, as shown by the recent shift in Exxon’s positioning on climate change emissions impulsed by their shareholders. 

→ Download ESG and Collaborative Data Science: A Necessary Bet for Financial  Institutions

But commitments are no easy things to take. All financial organizations are well aware that words have an impact and that scrutiny on alignment between commitments and actions is a must-do. Now that decisions are taken, the next step requires the firm acceleration of business processes revisiting — where financial analysts play an essential role.

What do financial analysts do? They assess the risks and opportunities supporting financial decisions. Whether they work within banks, insurance companies, asset managers or rating agencies, they are at the forefront of ESG integration in these decisions.

ESG cycle

They of course work tightly with ESG experts. But in a highly competitive market where financial margins are closely watched, no organization can afford to double their staff to handle ESG, nor would it drive successful integration of ESG into financial decisions. This leaves only one course of action: empowering the financial analysts with the capacity to embed ESG in their day-to-day processes. 

Practically speaking, what does it mean? Financial analysts need to be able to integrate, in their work, a variety of ESG signals coming from a broad range of sources, which we’ve outlined below: 

1. ESG Metrics Produced by External Providers:

From generalists offering a broad range of raw KPIs and scores such as Sustainalytics to specialized players such as Carbon 4 Finance and alternative data suppliers such as Kayrros, the financial analyst has to navigate through a broad range of signals and digest them into understandable insights. The financial analyst needs to be able to easily drill into these sources of data, understand their evolutions and underlying factors, and fuel these into his or her decision making. Ensuring financial analysts have a simplified, central access point to data sources selected by their internal ESG experts, with capacity to blend, drill down into specifics, and integrate with financial criteria is critical to accelerating ESG embedding.

APPL internal ESG ratings

Example of a visualization of Sustainalytics and market data built in Dataiku, broken by sector.

Practically speaking, here’s an example of what it can mean: As an oil and gas industry expert, I want to understand the CO2 emissions (scope 1, 2, and 3) of all main European players and I will use Carbon 4 Finance to do so. But I also want to integrate a precise measurement of methane emissions, provided by Kayrros, and apply a malus on any environmental and social controversy. I thus need to be able to blend all these insights with a weighting which can be unique to my decision process. My final objective is to weigh in these different factors as I issue my recommendation on future financial performance of these different players and the sector they form. 

external providers of ESG data

2. ESG Insights From Unstructured Data:

Numbers are numbers. However, they are only as good as numbers can be in an area which is far from being supported by global-enforced norms. For financial analysts, confronting KPIs to facts and insights from documents, public information, newsfeeds, and more is paramount to balance perspective and make the best informed recommendations.

Leveraging data science to accelerate this information digestion is key. As an illustration, Dataiku’s interactive document intelligence application combining OCR and sentiment analysis enables financial analysts to easily scan through documents and quickly understand key insights and outputs. 

Dataiku’s Interactive Intelligent Application in practice, applied to ESG

Dataiku’s Interactive Intelligence Application in practice, applied to ESG

When put into action, it means that the financial analyst doing the above study on energy companies will be able to easily scan through these companies’ reports, reports produced by other internal or external analysts, and NGO reports to balance out his or her perspective on each company and understand the detail of their actions.

3. From Individual ESG Insights to Integrated ESG Overview:

When analyzing issuers or assets, financial analysts have to absorb more than ESG metrics. They have to blend these together, apply specific weights and, more importantly, integrate these with financial information to create differentiating insights. They notably have to precisely understand how ESG performance has weighted on past company or asset performance, and integrate this in their forecasting models.

Now, of course they don’t do this alone. They are supported by a broad value chain of contributors made up of ESG experts, quants, and more, with much tailoring to be done to adapt to the specificity of each process, industry, and asset class. Agility in building these unique models, through collaboration, with capacity to govern resulting data projects becomes essential to successfully empower the financial analyst with ESG.

Supporting This Transformation With Dataiku

Dataiku’s centralized analytics and data science platform is uniquely positioned to support financial analysts through this ESG journey. Applications are broad, ranging from: 

  • leveraging advanced data science projects packaged in user-friendly applications
  • significantly accelerating data ingestion from multiple data sources and data types to develop unique models
  • easily deep diving into metrics, understanding models built by other teams, and reusing elements from previous projects
  • all in a fully auditable and transparent manner, rooted in collaboration 

As finance firmly engages itself in this climate change transition, acceleration and capacity to impact all processes will be key. While this blog post focuses on ESG for financial analysts, needs are of course much broader and will demand having the capacity to leverage a centralized framework while tailoring analytics to the specificity of each process, in a rapidly evolving manner. Embracing a platform approach will act as a key catalyst for financial players to emerge as winners of the ESG race, and contribute to the shaping of a more sustainable economic environment. 

You May Also Like

Dataiku Makes Machine Learning Accessible, Transparent, & Universal

Read More

Explainable AI in Practice (In Plain English!)

Read More

Secure and Scalable Enterprise AI: TitanML & the Dataiku LLM Mesh

Read More

Slalom & Dataiku: Building the LLM Factory

Read More