WTW: Radar delivers powerful analysis and insights
Serhat Guven, managing director and global proposition leader for personal lines pricing product claims and underwriting at WTW, explains how future-focused pricing, claims and underwriting starts with its Radar platform
How has WTW helped insurers with their main challenges in the last year?
A main challenge for insurers is to develop sophisticated customised products that meet and exceed consumer expectations. WTW’s Radar is tailored for the insurance sector by insurance experts to overcome challenges in creating the right rate for the risk in a timely manner. This is done by providing an analytics environment that is automated, intuitive, comprehensive and well governed.
Radar’s ability to implement predictive machine learning models into clients’ operations at the touch of a button, via a governed process, facilitates massive pricing agility and ensures 100% accuracy of the rating process. As such, the time from data analysis to model deployment is extremely short, but the quality and accuracy of the analysis remains very high. Our typical feedback is that Radar contributes to a significant reduction in combined ratios, given the significant uptick in agility, accuracy and insights it offers.
What technology trends do you expect to shape the insurance industry in the near future?
The demands on technology from carriers and customers are increasing exponentially. The market is forcing carriers to create more personalised and bespoke insurance products as quickly as possible. The technology required to deliver this has to be flexible enough for the changing market conditions, automated for efficiency and speed to market, and well governed.
Insures want fast, efficient deployment of complex rating algorithms that do not require the inefficient re-keying of rates. In addition, insurers demand the technology provides the ability to accurately test the impact of rates and predictive models in an off-line, real world context prior to deployment, but also to easily monitor the performance of models once implemented.
How do you expect risk modelling to change over the next five years?
Pricing is absolutely central to the success of insurers in all lines of business and, as personal lines markets become more competitive and commercial lines pricing becomes more sophisticated, the need for accurate, agile and governed analytics is increasingly in demand.
As we have seen recently, the availability of more data and the sophistication of traditional machine learning algorithms have created increased concerns about transparency among insurers, consumers and regulators. Recent updates in WTW’s Radar suite overcome this by introducing a ‘market-first’ capability that makes it significantly easier to benefit from the full predictive power of traditional machine learning, without losing transparency.
How will insurers’ stress and scenario testing practices change in the future?
Stress and scenario testing are fundamental to insurance operations. While there will always be a need to have an interactive environment where it is easy to perform ad hoc testing, there is now a greater recognition that there are millions upon millions of potential scenarios. This is where automation and speed play a critical part of the process. WTW’s Radar solution allows clients to generate literally millions of options and then using a customisable and sophisticated search algorithm, identifies appropriate scenarios that align with objectives.
What technology improvements has WTW made in the past year? Will there be further innovations this year?
WTW’s Radar Suite is on a quarterly release cycle that provides updates along five key themes:
- Analytics – specifically delivering enhanced analytic capabilities that is targeted to the insurance industry. For example, transparent machine learning algorithms.
- Integration – Connecting with wider systems, such as more seamless integration with open-source environments
- Enhanced speed and scale – both in terms of underlying architecture environments and optimising efficiency in statistical algorithms
- Enhance user experience – making it easier and more efficient for users to access and customise solutions
- Underwriter accessibility – delivering contextual insight to underwriters to enhance performanc
- We maintain an active roadmap of future enhancements that is reviewed based on emerging market needs and client feedback. These updates are made available to existing clients to accelerate their pricing and underwriting capabilities
- We continue to invest significantly in Radar to deliver to ongoing value to clients and continue to meet their current and future needs.
