January 15, 2014

Farm Bureau Insurance of Michigan wanted to offer its customers more self-service online while preserving the agent's role as a trusted advisor. However, its legacy systems presented several barriers.

Farm Bureau, based in Lansing, Mich., is a multiline carrier with $709 million in property-casualty and life premium. The insurer wanted to offer customers an intuitive agent locator, online quoting, policy inquiry, ability to request changes, access to documents, electronic bill payment, claims reporting and status inquiry -- all while keeping the agent informed.

Its executives wanted its customers to be able to go on its website to answer questions such as 'What's my deductible?' 'When is my payment due?' 'When is my policy up for renewal?' 'What's the status of my claim?'

A prerequisite to all of this was overcoming four key data challenges. First, customer data was fragmented across multiple source systems. There were separate policy administration systems for auto, home and life. Developing the common notion of a customer was not straightforward. Previous attempts to piece data together from disparate sources had led to "integration spaghetti."

Cheri Barnhart, Farm Bureau Insurance of Michigan
Cheri Barnhart, Farm Bureau Insurance of Michigan

Second, there was inconsistency and incompatibility in data formats. There were systems from multiple vendors, and each vendor implemented a custom data representation, some of which were proprietary.

Third was varying data quality. Separate business units manage specific product lines and operate fairly autonomously. This has resulted in different data entry practices leading to inconsistencies.

Fourth, the systems were only available in defined windows of time during the day, not 24/7. Furthermore, updates made throughout the day were not available in the system until after-hours batch processing was completed.

To implement its online customer self-service roadmap, Farm Bureau IT staff chose insurance technology consulting firm X by 2. This included designing and implementing a strategy to address all four data challenges at the outset. The project team closely blended in-house and X by 2 teams, and included business analysis, testing, data developers and project management. Team building and having a close partnership were priorities from the start.

The strategy called for a holistic data management approach and program that would serve as the backbone for customer self-service and also serve as an enabler for service orientation, data warehousing, and core systems modernization -- projects scheduled for the near future.

The core of the solution involved creating an operational data store (ODS) with a canonical data model based on a reference model published by the Object Management Group (OMG), a nonprofit organization. OMG's reference model was chosen because it's a community-developed, open-source, and licensing-free model, which makes it a very suitable starting point for further development. Over the years it has received increasing acceptance in the insurance industry.

The ODS would serve as a central repository where data from all of the source systems relevant for customer self-service could be consolidated. With all customer-related data in a single repository, forming the previously elusive holistic view of "the customer" started to become possible. By exposing the data in the ODS through data services using the OData standard, this holistic view of "the customer" was made accessible to other applications.

[From X by 2's Frank Petersmark: What's so innovative about innovation?]

The three-year project is ongoing, but Farm Bureau Insurance has already achieved many of its goals, including 24/7 availability, online access to policy details and documents, improved online payment capability with comprehensive bill presentment, expanded payment options and increased ease of use, and online first notice of loss (FNOL) claims reporting. Along the way, the team learned these key lessons:

  • Enterprise assets are best built as a means to an end. The ODS was built to enable customer self-service, and an ETL framework was built to aid development. Reusability of these assets for other enterprise initiatives was a purely opportunistic benefit.
  • Think big, start small with enterprise architecture. While the ODS was meant to enable delivery of customer self-service, its role as a stepping stone towards service orientation, data warehousing and core systems modernization was contemplated from the beginning.
  • There are no silver bullets, only skilled marksmen in successful implementations. Investing in team-building from the beginning was invaluable.
  • Build in quality from the beginning by following iterative development and testing.
  • Challenge the status quo. Trailblazers walk a lonesome path, but crowds do come after the chasm's been crossed.

Farm Bureau's successful data management program has spanned several business and IT disciplines, and is a good working example of the power of focused collaboration and shared business goals and objectives. The key to ensuring outcome coherency across all of the business and IT disciplines was the early adoption of a practical and sustainable architecture, including the use of the OMG data model.

For Farm Bureau, it was crucial to have an architect capable of conceiving the initial infrastructure, data, and integration architectures required for the successful implementation of the project. Beyond that, the enterprise approach taken by the team will allow Farm Bureau to leverage its modernized data management program to deliver improved customer service for years to come.

About the authors: Samir Ahmed is an architect with X by 2, a technology company specializing in software and data architecture and transformation projects for the insurance industry. Cheri Barnhart is director of information systems at Farm Bureau Insurance of Michigan overseeing Web, architecture, and data services.