Company profile
Euler Hermes is a global provider in trade credit insurance, surety bonds, commercial debt collection, and more. Our services enable companies to choose the right customers and markets, protecting their financial health and cash flow from bad debt losses. We empower businesses to trade with confidence through: - Knowledge: Our worldwide research team delivers economic intelligence, market insights and industry risk analysis that provide valuable insight into the financial stability of the companies our clients do business with. - Growth Acceleration: We help clients avoid catastrophic losses and grow profitability by minimizing commercial and political risks encountered in trading. - Peace of Mind: We actively monitor our clients’ accounts to provide advance warning of potential threats and handle claims and collections so our clients can stay focused on their business. Euler Hermes, a subsidiary of Allianz, is listed on Euronext Paris (ELE.PA). We are rated AA by Standard & Poor’s.
Division: Euler Hermes – North America
Industry: Insurance
Year: 2019
Student Team
Ruben Alba
Project Role: Business Strategist
Education Background: Bachelor, Electronic Engineering, Julio Garavito Colombian School of Engineering; GMBA, Thunderbird School of Global Management (USA);
Younes Amor
Project Role: Solution Architect
Education Background: B.Sc., Computer Science, Al Akhawayn University
Tamar Bulka
Project Role: Modeler
Education Background: B.Sc., Computer Science, McGill University
Cindy Qu
Project Role: UX/UI
Education Background: B.Sc., Chemistry, UBC (University of British Columbia)
Project description
In this project, student collaborated with the Credit Insurance team on developing a solution extract data from financial reports in a more automated fashion and investigating risk assessment factors The provided recommendation was created standard report readers and utilize text analytics to leverage new sources of risk factors to increase Coverages and decrease Net Loss Ratios.
Pain Points
- Fairly manual process on the data extraction phase
- Utilizing traditional risk factors based on past structured data
Data Inputs
- Financial Statement reports
- Dun & Bradstreet reports, history, comments
- Historical Risk Gradings
Analytic Methodologies
- Leveraging sentiment analysis on public data (i.e. analyst reports) to look for risk factor applicability and correlations
- Use of Gunning Fog Scores for Sector and industry grading
- Testing and Model validation using Logistic Regression, SVM and Random Forest algorithms
Technical Solutions Developed
- Data sources included 10K filings and Twitter analysis
- Text extraction involved identifying scanned PDFs, correcting for skewness, extracting text from images, and placement of correct text using ML algorithms on the extracted words
Partner Commentary
Partnering with McGill University’s MMA program was a great way for us to magnify resources and explore novel new concepts for the company. Like all businesses, limitations in resources often leave experimental ideas and processes unexplored. Our student cohort was not only able to provide key technical aspects we have incorporated into our business processes, but also prototypes for evaluating performance in our industry. In addition to the tangible outputs, we hired one of our students within days of their graduation.
—Aaron Lindstrom - Regional Head of Transformation and Digital Partnerships, Americas Region, Euler Hermes
Student Commentary
Cindy Qu – UX/UI
For people with a few years of work experience, the direct collaboration with the client during the capstone project not only provided a supportive environment to take my current experience and skillsets to another level but also allowed me to develop my leadership skills in a team environment with analytics context which was exactly what I came to the program for.
The faculty and client mentorship during the program absolutely exceeded expectations in many ways. When we ran into a wall during the capstone project due to limited knowledge in the analytics field, the professors steered us to inspiring and helpful topics and publications which we managed to incorporate into our final deliverables and impressed the client; the workshops organized for the program involved high-profile companies and the hottest topics in the analytics field and allowed me to easily connect with potential employers which I was interested in. Last but not the least, the MMA staff supported our class like a family would – they were responsive at all times, cared about each individual’s concerns and needs, and always up for a chat with or without a drink. Not a single day was the same during the MMA program, and the faculty made sure we were fully supported and highly engaged throughout the program!
Ruben Alba - Business Strategist
The most exciting part of the project was leading a diverse team with different cultural and professional backgrounds without any experience in the Insurance Industry to become a High-Performance team. The solution proposed exceeded the initial goals and expectations of our client; as a result, Allianz – Euler Hermes is implementing the solution and hired one of our team members.
Allianz – Euler Hermes asked us to automate the extraction of information from financial documents. We took a deep dive approach to our client's business to fulfill their needs in a simple but effective way by using an Expert System that automated the reading of financial documents and complemented it with non-traditional data to improve the credit risk scoring process accuracy. Our solution reduced the time credit risk analyst spend in non-adding value tasks and potentially can increase the net profits between 4%-7% on yearly bases. Providing a technical solution with business impact is what the MMA is about.
We faced three main challenges: The first one was understanding the Insurance industry and EH’s business objectives. The second was to craft a solution that addressed our client's needs while having an impact on their business. In both cases, team engagement and collaboration from our different areas of expertise were key to succeeding — each member was able to contribute to the project. Finally, gathering the right dataset to prove our solution was one of the most difficult tasks and it is an example of how difficult is to get/build a proper dataset in a real-life environment.
It was amazing to collaborate with C-Level and senior members of Allianz-Euler Hermes. They were supportive of our initiatives and provided us with the necessary elements to work from day one.
In terms of career services and networking events provided by the MMA program and Desautels, they are critical to have a soft landing on the Canadian job market. They help us to develop our networking and interviewing skills and tell us the particularities of how recruiters work in Canada.
My advice to future MMA students is to focus on the experiential opportunities as this is what will craft your technical, business and leadership capabilities. Good grades are great but successful implementation of a real-life project is what will make you stand out from other candidates when you are entering the job market.
