[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries

IT: Closing the decision technology gap

In today’s fiercely competitive insurance industry, challenging accepted best practices can be the key to success. Through my discussions with insurers across Europe, I have found four areas where leveraging innovative technologies can make a huge difference for actuaries to improve the decisions made on every policy, every claim and every customer.

1. Connecting Decisions
Most insurers have several systems in place for policy administration, focused on granular customer and transaction details. This data is just what customer-facing employees need, and the details can be a gold mine for actuaries looking for risk patterns. But the actuarial team may find itself spending months retrieving this data for analysis, and it can be nearly impossible to tie together data scattered across organisational silos in order to drive customer-level decisions.

For example, claims departments need to collect information from various parts of the organisation, from underwriting to payments. The more intelligence available, the better the chance to identify claims patterns, detect fraud, and find opportunities to increase profitability. And once that analysis is done, insurers need to connect the decisions made across the organisation, which isn’t common practice, even among the biggest insurers. Business rules management systems provide the shared decision management infrastructure needed to improve risk management, marketing and customer retention.

2. Factoring macro-economic forecasts into customer risk analytics
Because the economic crisis has changed the way customers behave, past customer behaviour is not as strong a predictor for future behaviour as it was. This means our whole approach on risk assessment needs to change; we need to understand how changes in the economy impact customer behaviour.

There are new analytic methods for forecasting how macro-economic conditions, such as unemployment rates, are likely to impact customer behaviour, such as the number of claims filed. We have studied the relationships between up to 150 different economic indicators and consumer risks, then used these relationships to build more forward-looking predictive analytics.

This kind of analysis can not only help insurers calibrate models — for example, to understand how claims loss ratios will change at different score thresholds — but also revise and improve claims management and underwriting decision strategies. Insurers can select the economic forecast they believe is probable, then use the associated index metrics to adjust their score cut offs for policy approval. In this way, it is possible to maintain a fairly consistent claims and fraud rate across changing economic conditions and evaluate alternative designs or policies without having to wait to see how they play out in the real world.

3. Making Real Time Decisions
Real-time decision making is especially beneficial for the underwriting department as it allows automated enrolment. Aviva Health, for instance, reduced enrolment time from 22 days to 6 minutes with 70 per cent of applications being processed without manual intervention. Business rules and predictive models can speedily evaluate individual applications, delivering a quantitative measure of risk and improving efficiency.

This kind of acceleration is also critical to improving claims fraud detection. Predictive models that instantly analyse claims, prior to payment, have helped insurers identify as much as 50 per cent more fraud with a considerable decrease in false positives. Using predictive analytics combined with traditional rules-based systems, insurers can locate aberrations in customer claims more quickly and more accurately. The speed at which claims can be processed will also increase, as the analytics can easily identify which claims should be paid automatically and which should be reviewed.

Business rules can also help actuaries align reserves with claims. Rules can instantly flag discrepancies between the estimated and actual claim. If a claim that initially looked like it would be £50,000 turns out to be £500,000, the actuary can spot this discrepancy sooner and take steps to amend reserves as appropriate.

4. Optimising Customer Lifetime Value
Rather than focusing on how profitable a given customer is for a specific policy, insurers can and should optimise each customer’s policy portfolio – home contents, building, automotive etc. – based on that customer’s needs and risk.

Analysing customer data from the bottom up, ensuring an understanding of their complete range of requirements and preferences is built, will enable insurers to optimise decisions about which offers will work best for an individual customer. Actuaries can use existing risk models with psychological approaches like the Myers-Briggs personality measure to establish a “responsibility score” for each individual. Optimisation technology can then be used to determine the ideal portfolio and offers for each customer.

These uses of decision management technology — business rules, predictive analytics and optimisation — have been successfully deployed in the banking industry. It’s commonly said that the insurance industry lags bankers’ use of technology by 10 years. Adopting the approaches in this article can help insurers close the gap.


Larry Jacobson, Senior Consultant and Insurance Specialist at FICO