Generalised linear models (GLM) have been used in the industry to predict frequency and severity of claims for a very long time. The modelling and maintenance process has usually been effort intensive due to manual test and learn necessities, therefore time-to-market of changes has been slow. Democratisation of machine learning algorithms has opened up new possibilities to develop new powerful models with faster time-to-market. With the new approach external data is extensively used non-linear patterns in data are captured unlike GLM approach. We design and develop the pricing models for our clients to create competitive advantage in the market. Our engagements typically go beyond modelling. We design the overall structure including all layers of pricing (technical and commercial) and the process to make quotation acceptance rate at optimal level.
Machine Learning & AI Teams
Past market examples show that profitability levels can drastically change with optimising tariffs without negative effect on growth. We help insurers determine the ideal coverage scheme and underwriting rules based on their customer portfolio using descriptive methods. Our approaches will be reusable by our clients' teams thanks to complete transfer of our knowledge.
Market dynamics require the insurers to bring new ideas quickly and combine different products or coverages into specific value propositions. We help our clients pinpoint customer needs, make necessary combination of coverages at the right profitability levels and determine the go-to-market structure required for successful market entry.