Esurance Puts Analytics Closer To The Customer - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
IT Leadership // CIO Insights & Innovation
News
4/28/2015
11:50 PM
Connect Directly
Twitter
RSS
E-Mail
100%
0%

Esurance Puts Analytics Closer To The Customer

A new rules-engine platform lets the insurance company offer more personalized packages when selling policies online. One of eight Elite 100 Business Innovation award winners.

[Editor's note: This article has been updated to reflect a clarification by Esurance.]

Esurance, a pioneer that started selling auto insurance online 14 years ago, is finding there are ways to make the online experience a closer equivalent to sitting down with a knowledgeable insurance agent.

"For most customers, it's complicated choosing the right insurance," says Tolithia Kornweibel, Esurance VP of analytics. Esurance needed a platform that could quickly make insurance knowledge available to online customers as they made their choices. The result was the Coverage Rules Engine, a system that captures knowledge of the insurance business -- for example, what it will cost to guarantee replacement of a car or other goods -- and makes it available to customers on a one-to-one basis.

Esurance's Coverage Rules Engine platform started with the open source Drools business rules engine, which the company customized to process Esurance information. Esurance has three staffers who collect business intelligence with which to build rules, which they do by combing through queries against the historical data in its data warehouse. The researchers also spend time talking to Esurance's agents, who help customers online and on the phone. The result: the CRE can produce 8 billion different combinations of coverage.

One eye-opener they spotted, for example, was that insurance agents in face-to-face discussions often concluded with a suggestion to add towing and roadside emergency insurance. Customers frequently signed up at that point. But Esurance customers, browsing through options on their own, seldom added towing.

The rules researchers, headed by Benjamin Wolfe, manager of data science, shared the finding with Kornweibel. "Towing insurance doesn't cost very much. We could see the pattern of agents selling it a high rate," says Kornweibel. "Why not online?"

Tolithia Kornweibel, VP analytics, Esurance

(Image: Esurance)

Tolithia Kornweibel, VP analytics, Esurance

(Image: Esurance)

In early 2014 Esurance set up an experiment. Developers added a button to the right hand side of page, asking customers, after they had selected other options, whether the customer wanted to look at towing insurance options. If they didn't, they simply clicked through to complete their deal. But as with in-person sales, many decided they wanted it, and towing insurance started getting added to policies on a regular basis.

New Opportunities

Many online car insurance customers are also apartment renters, so in 2014 Esurance started showing a rental insurance option to those customers, thus creating a new vein of revenue. It's offering home owner insurance also, even though the Rules Engine needed to gain a lot of understanding through new rules to determine what a prospective customer's home was worth and how much it would cost to insure. Home values have a lot to do with what's going on with values in a given neighborhood.

Esurance spent about 18 months building the Coverage Rules Engine system. It started in 2013 and put the system into production use in 2014. The first goal was to make it easier and faster for Wolfe and the data researchers to put their findings into rules governing coverage.

But that faster analysis translated into richer customer interaction. Indeed, the CRE platform brings the world of online insurance closer to the traditional world of the insurance agent sitting down with the customer face to face. Data analysis from the data warehouse can help formulate the packages of options that are most likely to succeed, but "we never assume we're getting the answer right," Kornweibel says.

The CRE relies on the answers that a customer provides for 20 to 25 questions, which provide about 55 variables that it can work with in modifying initial package presentation. The CRE "is creating models and rule sets to drive personalization in real-time," Kornweibel says.

The performance is constantly measured by comparing the success rate of customers signing up versus those who leave without buying.

Agility is a key piece of the platform, in order to allow for targeted experiments like the one for adding a towing option. The data research team can quickly add those experiments by composing rules in R, rather than needing to go through a much longer IT development process. (R is a language frequently used in data mining and statistical research.)

By implementing the Coverage Rules Engine platform, Esurance has saved the equivalent of $500,000 to $1 million a year in IT development or analytics software-as-a-service subscriptions, Kornweibel estimates. The CRE itself cost less than $1 million to develop and bring into production.

The company also thinks CRE's customer-facing function helps Esurance customers buy the right amount of insurance coverage.

With the homeowner's option, for example, many customers will click through it because they already have insurance and aren't interested. But those who try it out find they can qualify for a discount by combining home and auto with one insurance provider.

That's part of the purpose of the CRE. The engine knows what's possible behind the scenes with insurance combinations, and it's bringing that information forward in quotes for the customer. Esurance advertises it will personalize your car insurance in half the 15 minutes that another online insurer (GEICO) boasts about, and that can build customer loyalty. "We wanted to keep them engaged by giving them the right tools," Kornweibel says.

Charles Babcock is an editor-at-large for InformationWeek and author of Management Strategies for the Cloud Revolution, a McGraw-Hill book. He is the former editor-in-chief of Digital News, former software editor of Computerworld and former technology editor of Interactive ... View Full Bio

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
tzubair
50%
50%
tzubair,
User Rank: Ninja
4/29/2015 | 9:03:01 PM
Data independence
"The data research team can quickly add those experiments by composing rules in R, rather than needing to go through a much longer IT development process"

I think that's a big step on its own. If the business users can themselves become tech-savvy and independent, it would greatly increase their productivity. You don't have to be dependent on the IT guys anymore and wait for them to give you reports. I think other organizations should encourage this behavior.
Commentary
Enterprise Guide to Edge Computing
Cathleen Gagne, Managing Editor, InformationWeek,  10/15/2019
News
Rethinking IT: Tech Investments that Drive Business Growth
Jessica Davis, Senior Editor, Enterprise Apps,  10/3/2019
Slideshows
IT Careers: 12 Job Skills in Demand for 2020
Cynthia Harvey, Freelance Journalist, InformationWeek,  10/1/2019
White Papers
Register for InformationWeek Newsletters
Video
Current Issue
Getting Started With Emerging Technologies
Looking to help your enterprise IT team ease the stress of putting new/emerging technologies such as AI, machine learning and IoT to work for their organizations? There are a few ways to get off on the right foot. In this report we share some expert advice on how to approach some of these seemingly daunting tech challenges.
Slideshows
Flash Poll