Thematic portfolio construction for a private bank

A leading Swiss private bank needed to automate and enhance the selection of ETF’s aligned with specific investment themes, previous attempts failed due to biased data (greenwashing, languages). The objective of the project was to compute purity scores of ETF’s (and its constituents) against a defined theme.
Mathematical modeling
Financial Services


Constructing a thematic investment portfolio is becoming a challenge as the number of companies grows and as they diversify their activities. Portfolio managers are also faced with more and more competition and need to remain on the edge and provide their customers with more tailored products. Our client was a renowned bank that wanted to approach the thematic investment problem with a highly granular approach: as industries become more complex, so too should the way we assign companies to fields of activity.

Companies tend to be active in several sectors and can hardly be classified into a single category. Consequently, a company might be exposed to different topics with different weights depending on how much of their activity is focused on that topic. Clients want a more granular and intuitive approach to thematic investing. Investing in “Pharma” is no longer granular enough, as clients want to invest in themes such as “Immunology” or “Oncology”.


We provided our client with a scoring methodology that enabled scoring the exposure of any company to highly granular topics. This solution not only provides a new way of investing in highly granular topics, but also enables investors to evaluate how exposed to other themes a thematic portfolio really is. In turn, this can lead to hedging with respect to various themes.


Azure Cloud
Power BI


Understanding the problem was a key part of this project, and Effixis’ background with financial projects enabled our team to quickly move on to the implementation phase. Along with our client, we defined an ontology that described through what themes they wanted to see in their investment universe. In doing so, we defined a hierarchy of categories that ranged from “health” down to the manufacturing of individual vaccine components. Effixis’ state-of-the-art NLP technology was leveraged to:

  • Gather a large amount of data on companies within the client’s investment universe.
  • Evaluate within this newly created dataset how each company was active in all the fields of a pre-defined ontology.

Similar case studies


Cyberdefence & Language Models – Enhanced monitoring

Innovative AI powered dashboard to foster cybersecurity news monitoring.

Revolutionizing legal document research

AI-Powered semantic search and novelty detection for efficient document analysis.
Financial Services

A new approach to the Swiss Solvency Test (SST)

A case study on delivering a groundbreaking open-source solution for FINMA.

Upskilling 1,000+ Managers in Digital Technologies

A Comprehensive training program in Digital Foundations, AI, and Data Ethics.

Global monitoring of specialized chemical industry 

Leveraging our NLP-based Intelligent Tagging Platform in the chemical industry.
Real Estate

Structured feature extraction from real estate listings

Extracting structured real estate information from text using cutting-edge NLP.


Stay tuned, get inspired

Join our newsletter by filling out the form below

Stay tuned !​

Don’t miss out on our latest news – subscribe to our newsletter today!