Insurers have begun to embrace the onslaught of data in a variety of formats - e-mail, PDFs, photographs and social media feeds, to name a few. The idea is to turn the vast menagerie of unstructured data into useful business insight, with the goal of providing better results to clients while boosting sales. But achieving this end poses both a business and a technological challenge that few have been able to tame so far.
"Insurance companies are data-rich sources of all kinds of new and existing information that lends itself well to these kinds of analytical tools," says David Castellani, SVP of business information at New York Life. Tapping into unstructured data to glean useful business insights, he adds, is a new but fast-growing trend. "This trend is nascent, but it has huge implications for our industry in general."
Insurers traditionally have depended on document management systems for storage, management and retrieval of key business forms, including customer and claims information. While these systems continue to house the industry's essential business information, many large insurers are trying to leverage the data contained in a host of new media types to help boost the business.
Among the top document management and workflow priorities identified by research firm Novarica in its 2013 report, "Business and Technology Trends: Commercial Lines," were the ability to store and manage multiple media types, and turn this unstructured data into useful business intelligence.
Proponents of big data assert that through the use of predictive analytics, management can make better-informed business decisions through a combination of data analysis and scientific reasoning, rather than intuition or opinion.
"Many insurers are just beginning to leverage big data," says Martina Conlon, a principal in the insurance group at Novarica. "Large insurers are leading the way with collection, analysis and predictive analytics that incorporate big data, while smaller insurers are in more of an investigation and planning mode. Large direct writers are actively embracing big data and gaining insights on their prospects, customers, and risks."
One company that has managed to tame the wild data beast is GMF Vie, a subsidiary of GMF Insurance, a marketer of retail life insurance products in France with more than 800,000 customers. GMF Vie has realized a 200-percent to 300-percent increase in conversion rates for marketing campaigns targeting the top 10-percent highest-ranked customers identified by predictive models. The company uses predictive analytics software from KXEN, a predictive software firm that was acquired by enterprise software giant SAP in 2013.
Big data analytics may be most useful when applied in response to specific business questions. Thus, instead of using a single propensity model to determine if an individual is likely to respond to a marketing campaign, GMF Vie combines various models to address specific business questions, taking into account if the targeted customers have sufficient financial capacity, whether they are likely to churn in the next three months, whether they are expected to move their homes soon, or whether they are more receptive to offers by phone, text or e-mail.
GMF Vie combines vast amounts of big data on customers with third-party data, such as geographical marketing and demographic data. The company's data mining team then consolidates this information into project-specific data marts using up to 1,500 variables describing customers and their behaviors. It uses KXEN's InfiniteInsight predictive analytics to analyze the data and build customer segmentations and predictive models. Using KXEN to speed the data modeling process, insurers can more quickly leverage the ability to solve a problem, such as identifying which customers are most likely to respond positively to a marketing campaign.
The end result is that for existing high-potential customers, when marketing an up-sell offer, GMF is able to convert between 3 percent and 10 percent of targeted customers, a rate exceeding industry benchmarks.
GMF Vie is able to personalize each and every interaction to the individual, the channel, their interests, and their behaviors, says Christine Dalle, head of business intelligence and customer insight. She said the system is helping drive significant returns on customer analytics investments and optimize annual customer solicitation initiatives.
Another insurer that has embraced the big data challenge is New York Life. Castellani, whose role at the insurer is to bridge the company's business and IT functions, says the company uses a variety of tools as part of its big data strategy. These include data virtualization software from Delphix; Apache Hadoop for processing large datasets; and Splunk for visualizing and analyzing the new data.
"We want to make sure there is uniformity in how the data is extracted, scrubbed, catalogued and governed," Castellani points out. "This is the hard part in a mature organization with lots of data that is difficult to access."
TURNING TO SOCIAL
New York Life also is working on a pilot project with Hearsay Social, incorporating data from multiple online sources including LinkedIn, Foursquare and Facebook. Castellani says there are numerous potential applications for harnessing this kind of unstructured data to help improve the business. "For example, we have an active agency with 13,000 captive agents who work for the company, and we could use the data to develop a hiring profile that is more successful."
Another potentially beneficial area is in underwriting. "Insurers want to take on business that is profitable, and data analytics can point us to customers who are right for New York Life," he says. "The result is that you lower your churn" or customer turnover. Big data also can enable insurers to keep tabs on customer's wellness habits, or lack thereof. In other words, being able to maintain an insured population that exhibits certain habits, including regular exercise, proper diet and attention to managing stress effectively all contribute to what Castellani says "is a better set of clients for the insurer."
Of course, even basic business processes are affected by the need to embrace different data formats. At ACE USA, for instance, the company's risk management group had grappled with what Matt Merna, division president of ACE Risk Management, describes as "an old, archaic, manual process with paper forms. The clients had to sign 320 pages of documents," he adds.
The company sought clients' feedback on how they could simplify and automate the documentation and submission process for commercial uninsured and underinsured motorists coverage elections, while including electronic signature capabilities, with the client's PIN providing signature authentication. "We spent a lot of time with our brokers and clients to get their input and buy-in to make sure we delivered an end product that worked for them," Merna explains.
Using the company's Worldview Portal, clients including risk managers and insured can now sign and submit all paperwork including the required e-signatures with the click of a button using the ACE Worldview portal. ACE's internal IT staff built the online platform featuring 22 questions, effectively replacing hundreds of pages of paperwork, in about 10 minutes. The Web-based platform provides a step-by-step process, with drop-down menus and checklists, so that risk managers can easily identify pre-qualifying customer information and appropriate risk levels to meet differing regulations in 50 states.
The ACE Worldview Portal stores the forms in a PDF format as searchable documents in the portal library, enabling clients to have realtime online access. By clicking on "document library" in the portal, clients can check on their policies and download them as PDF files. Next up on the agenda, Merna says, the company plans to broaden the use of the Worldview technology across other ACE USA divisions.
Doug Bartholomew is a freelance business writer based in Berkeley, Calif.
4 Commercial Lines Doc Mgt Priorities
Commercial lines insurers are heavy consumers of inbound documents, according to Novarica's Report "Business and Technology Trends: Commercial Lines." From photographs, loss runs, and financial statements to schedules of vehicles and buildings, storage of these assets has been a requirement for some time. Most commercial carriers have had some sort of document management solution in place for years and some are now looking at upgrading. According to Novarica, priorities in these upgrades include:
1. Ability to store and manage multiple media types
2. Integration with core assets, whether those are underwriting systems, claims systems or portals
3. Indexing capabilities, i.e., the ability to turn the unstructured data (contained in most documents) into structured data that can be used to feed reporting or business intelligence tools
4. Ease of administration, including the ability to manage users, roles and access privileges, workflow routing and other elements.
