Euler Hermes has been a world-leader in trade credit insurance (TCI) solutions for more than 100 years. Tools like AI and machine learning are taking the firm’s offering to a new level.
1. How is insurtech improving customer engagement and experience in the TCI space?
In addition to obtaining coverage against buyer insolvency or default, one of the biggest reasons customers come to Euler Hermes (EH) is because they’re looking for a credit management partner – someone they can rely upon to give them advice on credit risk and help prevent claims. As a result, we interact with our customers almost on an daily basis – not just in the event of a claim. This creates multiple touchpoints with our customers, and to ensure every touchpoint and experience is a great one, we need to closely observe these interactions and proactively gather feedback. The user feedback of today becomes the user experience of our customers tomorrow, just like today’s user experience is the result of yesterday’s customer feedback. Insurtech helps us satisfy the evolving needs of customers in a variety of ways, from speeding up underwriting decisions and driving efficiency in claims and collections through the use of AI and machine learning (ML), to helping us stay ever-more connected to our customers through real-time client platforms and APIs. As our customers become more digitally savvy, they increasing demand real-time interactions and self-service solutions. This trend will continue as Millennials become a majority of the customer demographic, and we are constantly enhancing our platforms to meet this need.
2. How may Insurtech influence the future role of the broker/agent?
This is the million dollar question for the industry. Insurance, at its core, is a people business, and brokers and agents will remain an integral part of the ecosystem. However, the need for the industry to transform digitally is undeniable, and this has been accelerated by COVID-19. I see insurtech as an enabler of a more seamless and engaging intermediary-customer relationship. Digital platforms will deliver faster service and a smoother, more efficient process, while the role of the broker or agent will evolve to be more focused on risk management and insurance consultancy. Intermediaries will help clients develop holistic risk management strategies, guide them in choosing the best coverage and provide them with valuable insights and peace of mind. Brokers and agents who are able to do all this while adapting to new digital ways of working will be very successful and remain invaluable.
3. What is the key to successfully applying AI and ML to business processes?
AI and ML already have, and will continue to have, a tremendous impact on both our internal processes and our external offering – and this is visible on both the top and bottom lines of our P&L. Successful AI and ML initiatives start with a business problem, align with business strategies and focus on customer value. You can’t make the mistake of force-fitting AI into the business. Instead, you must let the business needs and goals dictate sensible applications of AI. Ensuring a consistent and reliable inflow of quantitative and qualitative data is absolutely critical for predictive models. You also need to clearly define how you are going to measure the success of the model output and its application to the business. This process requires strong collaboration between business owners and technical experts.
4. How are tools such as AI and ML informing better underwriting decision-making?
TCI is a data-driven business. Providing timely information on buyers is vital to support our customers in their credit risk management. Over the last 100+ years, EH has built a huge database of buyer information. We predict the probability of default of all buyers in our database, grading each from 1-10. Our grading predictive model leverages AI and ML, giving us a distinct advantage in predicting potential failures, but also helping us identify good buyers so our customers maximize their opportunities. Underwriting decisions are not based solely on the buyer’s financial situation. We are now able to capture many other variables, including buyer-seller relationships, payment performance and sector outlook, meaning we are able to make not just quicker but also more holistic, informed decisions.
5. How are AI/ML and automation improving claims performance?
The biggest impact AI has in the claims sphere is in detecting and tackling fraud – from both customers or their buyers. This helps improve internal efficiency as our claim assessors are now able to focus more on claims that need their expertise. AI and ML are also driving efficiency in the claim recovery process by identifying files that are likely to be easier to recover. This allows our collection agents to concentrate their efforts and prioritize their actions accordingly, increasing recovery speed and collected amounts. This improved speed and efficiency ultimately results in a better claim experience for our end customers.
6. What are the biggest challenges you face with Asian credit risk data? How is technology helping overcome these challenges?
While other insurances lines such as property often work a lot with pictures, we mostly work with highly structured data such as financial statements or administrative information. This data feeds algorithms that extract valuable information on company performance. The low level of official insolvencies in Asia creates an additional layer of complexity that we need to build into our performing model. Rather than looking just for insolvencies, we look for other events, often based on local market practices, that indicate failure. Access to data is improving but managing and treating this data has becomes the new challenge.