Persuading low-risk clients with attractive premiums, and expanding markets with competitive rates: using predictive analytics, property insurers can forecast losses more accurately than before and thereby attain a clear competitive advantage. Banks also profit from big data analytics by planning campaigns more efficiently and improving their message to customers.
Determining individual loss risks for individual insureds is a much more challenging task than identifying risk groups for classic rate formation. Auto insurers, for example, have a need to set the insurance rates for young drivers in the most personal way possible. We took on this challenge for a German insurance company. What resulted was that the potential of this classic risk group could only be fully exploited if the insureds were not all treated the same. By defining a new rate for young drivers based on an individual loss forecast, the company was able to considerably improve its risk structure for auto insurance policies. The lower the loss risk forecast, the lower was the individual rate. "Speedsters" with a high chance of an accident were persuaded to not enter into a contract, by means of very high rate quotes. The loss rate sank, while sales volumes increased at the same time. The risk structure in this rate class was considerably optimized by the use of predictive analytics.
With this process, which recognizes statistically significant patterns and relationships in a fully automated manner, and which comes up with highly accurate forecasts, banks can also increase their efficiency in retail banking. We have been able to demonstrate this in phone-based campaign management. The starting situation was that existing customers of a financial institute were offered a deposit account with an attractive rate of interest. A call center called 40,000 existing customers of the bank. Without Blue Yonder forecasts, the average conversion rate was twelve percent.
In order to use the employees in the call center in the most effective manner, the bank, together with Blue Yonder, determined the exact probable conversion rate for each individual customer. Instead of "dialing the alphabet," the calls could be prioritized based on their chances of success. As a result, only the customers with a high probability of conversion were called. The efficiency of the direct marketing campaign was quadrupled in that way. With that action, the time required was reduced by about a person year. A cost savings of about 80 percent was attained in that way. The average conversion rate rose from twelve to about 40 percent.
Two of the many advantages of big data analytics can be seen here, in figures: Predictive analytics increases the efficiency in the message to customers and is a valuable instrument in determining optimal prices – two important application areas for financial service providers. Because customers can inform themselves at any time using the internet about rates and conditions today, they have become very price sensitive. They will often negatively view standard offers that are not adapted to their individual desires and current life situations.
The industry has had at its disposal data from various sources for customer acquisition and customer retention and that can be used to develop customized products. With predictive analytics, in addition to the "classic" methods, complex relationships between factors of influence and targets can be recognized and used: The same data thereby uncover valuable information. Enterprises thus boost their potential that up until then had not been used. Blue Yonder forecast solutions provide results in near real time and continuously adapt to changing market conditions. In a market that is always very informed and up to date, that can only be an advantage.