The history of the insurance industry is really a history of the application of information technology. The ability of insurers to gather, analyze and publish information has always been determined by the state-of-the-art technology available to communicate, store and organize information.
Insurers were early adopters of international couriers, telegraphs, telephones, typewriters, filing cabinets, tabulating machines and mainframe computing.
Unfortunately, the organizational structures of many insurers are still defined by this 20th century relationship to information. This 20th century information ecosystem is based on assumptions that no longer hold:
* Gathering extensive information about prospective risks or actual losses is labor-intensive and time-consuming.
* Providing information to prospects is labor-intensive and expensive, and people who need to work with the same cluster of information need to be physically collocated so they can share access to information stored in physical form.
* People need to work in offices so they can be connected to telecommunications systems.
* Organizations require large clerical staffs to manage information.
* The ability to discern the patterns in large masses of data requires years of specialized training.
Many insurers will recognize these assumptions as underlying some of their own current organizational structures: * Their staffs work all together in offices, so they can be physically proximate to information storage (paper files, computer systems) and channels (mail, land-line phones).
* They have large, low-skilled clerical staffs to process paper-based information, often located in primary market cities, where real estate is extremely expensive.
* They pay expensive commissions to agents to provide information to prospects, and to gather information from prospects and provide it to underwriters.
* They have underwriting and claims processes designed to support the laborious assembly of a case file, and the review of that case file by a human expert.
* They continue to use data sources and elements in underwriting and claims uncritically because they've always used them, rather than re-examining to determine whether these (or possibly other) data sources are truly predictive of results, because the cost of analysis is perceived to be high.
All of these practices and prejudices are the result of a certain stage of information technology - the 20th century. As information technology has advanced, few companies have really taken the opportunity to think about the implications of those advances, and question their business practices.
PROVING ASSUMPTIONS WRONG
Assumption No. 1: Gathering extensive information about prospective risks or actual losses is labor-intensive and time-consuming. Thanks to commercial data aggregators such as FICO, LexisNexis, Dun & Bradstreet and others, as well as increasing movements on the part of government to make more and more data available in structured electronic form, gathering information is faster and cheaper than ever, and only getting more so. Already, leading personal lines insurers such as Mayfield Village, Ohio-based Progressive have stopped asking prospects to provide their VIN numbers, instead pulling the information from DMV sources and asking the prospect only to verify the information. It is not too far-fetched to envision a near-future of one-question underwriting: "What's your social security number, and may I have your permission to access information about you from commercial and government sources?"
While some insurers have doubts about the quality of data from these aggregated sources, I have asked several audiences of P&C insurers lately if they believe that their agents provide higher quality data than the aggregators. No hands have gone up.
Assumption No. 2: Providing information to prospects is labor-intensive and expensive. When information is tied to paper or people, providing information to markets is expensive. When information can be broadcast and made accessible online, it is much cheaper. While intermediaries continue to provide value to most segments of the insurance market, that unique value as an information channel has diminished. Insurers need to re-evaluate the value of their intermediaries and their strategies for them.
Assumption No. 3: People who need to work with the same cluster of information need to be physically collocated so they can share access to information stored in physical form, and people need to work in offices so that they can be connected to telecommunications systems. The growth of Web-based computing and the explosion in both mobile and VoIP telecom makes these postulates obsolete. There may be other reasons to physically collocate staff, but insurers should make sure that they know why they are doing so, and what they hope to gain. They should balance this against the potential benefits of moving more staff to the field closer to customers and intermediaries.
Assumption No. 4: Organizations require large clerical staffs to manage information. Paper-based information processing requires a large staff to organize, file, retrieve, read and re-format/reproduce that information. Digital data eliminates much of this need. The first wave of Internet adoption by insurers produced many empty cubicles in new business groups as applications arrived electronically, and were routed straight into underwriter workflows.
Assumption No. 5: The ability to discern the patterns in large masses of data requires years of specialized training. While this is still true, the analytical tools now available change the nature of this kind of analysis. Insurers should take advantage of these tools as force multipliers of their most insightful analysts, and also to explore the opportunities of looking at risk factors that they may not have previously considered.
Insurer operating models were not carved on stone tablets. They were developed in response to the limits of the available information technology at the time.
Insurers should periodically ask themselves not just how they can do their current tasks better, faster and cheaper, but why they do them at all, and whether those reasons are still valid given the current state of information technology. They may well be, but not asking the questions ensures that insurers will miss opportunities to change the game in their favor.
Matthew Josefowicz is director of the insurance practice at Novarica, New York. He can be reached at mj@novarica.com.
(c) 2009 Insurance Networking News and SourceMedia, Inc. All Rights Reserved.
