A New Age for Old Systems

The new millennium has brought new tools to the old game of legacy systems management. There also has been an increased emphasis on strategic alignment between business and technology as the metaphorical wall between users and technology collapsed long ago for the more strategically oriented companies.

More than three-quarters of a decade ago, the beginning of the new millennium--with its associated worldwide fears of Y2K--came and went. The tremendous bubble in systems consulting associated first with replacing and then recognizing the reality of merely patching all those legacy applications has returned to the more mundane, less globally significant concept of enabling competitive advantage. Still, here we sit approaching the year 2010 with much of the industry continuing to process the predominance of in-force policies on systems built nearly 40 years ago.

In the years since Y2K was crossed without serious trauma to the industry, the focus on replacing legacy systems has remained to some extent in the spotlight. The game has stayed the same; only the reasons have changed. Having successfully traversed the millennial boundaries, the driving reasons for the replacement of legacy systems have returned to the standard, business-constrained pattern of inflexibility, a diminishing labor pool of expertise, reduced speed to market on new products, increased cost of compliance, service environment complexity, and unreasonable total cost of ownership (TCO).

Yet like the still vigorous mainframes that have survived predictions of death by PCs, minis, and the Internet, legacy applications remain a critical element within the foundational infrastructure of many successful companies. Even those that have moved on to incorporate new generations of software can be found with highly profitable blocks that are laboriously plodding along on a COBOL- or Assembler-based mainframe system that falls into the CFO, Life/70, LifeComm, PALLM, or original Vantage or Cypros A/P class of systems. Traditionally by nature a conservative industry, insurance companies have reason to approach broad-scale replacement of these technological dinosaurs with trepidation.

Greater results have been achieved using a combination of staging, which is the use of new technology for a new product or business line, and wrapping, which is putting new features around the perimeter of legacy applications. Both approaches avoid the high-risk elements of long projects, exorbitant expenses, technologies that become outdated quickly, and technology-driven solutions by focusing on business-specific, near-term deliverables that enhance the ability to create, distribute, deliver, and serve insurance products.

Staging new-product introductions has provided many carriers the opportunity to install new systems in a planned manner, allowing policy attrition and replacement to move the processing gradually to the new platform, although not entirely eliminating the need for the older platform in the foreseeable future. Conversely, straight-through processing (STP) represents an excellent example of a successful wrap approach many companies have taken to replacing the new-business portion of their legacy applications. In both cases, the focus has been on a specific business need driving the technical solution with the constraint of a manageable near-term planning horizon.

One major contributor to the ability to approach solutions in this manner has been the advent of service-oriented architecture (SOA). It represents an architectural concept built upon stand-alone, functionally specific modules of code that are reusable enterprisewide across a diversity of products and needs within the enterprise. Based on recent surveys of insurance company executives, we know SOA is perceived as a "game changer" for new business, underwriting, and service functions. In fact, it has been estimated the average insurer today has approximately 15 SOA services installed with about a half of those focused on new business and underwriting. Note these successes have been built based on a staged approach vs. a complete transformation of a legacy environment into an SOA-based one.

Taking a less technical approach, the advent of business process management (BPM) has opened many doors to improved efficiency and effectiveness for insurers. Here, the focus is less on technology and more on the concepts of process improvement borne of total quality management, lean manufacturing, and Six Sigma disciplines being applied to the business of insurance. For those companies willing to invest in the training and then to internalize the discipline and rigors inherent in BPM, there have been huge gains in efficiency, productivity, control, and agility.

BPM has been labeled by many as yet another "game changer" for the industry, one that does not depend upon tremendous technological investments in the replacement or enhancement of systems. For those choosing to benefit from this approach, keep in mind the three keys necessary for BPM efforts to be a success. They must be:

1. Incremental

a. Start small and grow to a big impact over time.

b. Build on a process of continuous small improvements.

c. Leverage the growing foundation of improvements for each new stage.

2. Measurable

a. Metrics and measurements must be incorporated into the process from the beginning.

b. The entire process must incorporate their automatic capture and tracking.

c. Each effort must produce a material and visible impact on performance.

3. Repeatable

a. There must be a compounding benefit building on the skills and competencies gained.

b. Future processes should be automated once the initial victories have optimized them.

c. There must be a gradual, duplicative spread of the new processes throughout the enterprise.

For the companies that have tried but failed to integrate BPM into their enterprise, the causes typically fall into some combination of five categories: missing the opportunity of repeatability, following in the path of least resistance, neglecting stakeholders in validating the designs, getting stuck in analysis paralysis (a common trap for our industry caused by an unwillingness to accept "good enough"), and selecting free or cheap BPM technology. While these may seem to be straightforward issues, their prevalence in failed efforts shows their insidious nature.

For those facing a portfolio of legacy applications and an increasingly demanding environment, the best advice is to develop a long-term, business-driven transition plan that is founded on a strategically determined infrastructure and servicing environment. Knowing where the road ends helps provide guidance along the way in making the appropriate decisions necessary to eventually complete the journey.

With this plan in place, let the business needs drive the technology decisions and solutions. Focus on transitions, wraps, and a staging of solutions in measurable and meaningful increments. Avoid the vainglorious overhauls and replacement projects, choosing measured steps instead. Ideally, forming appropriate alliances with external partners who have expertise in the various tools involved in your transition will prove beneficial; however, take care to validate references and avoid "vaporware," unrealistic promises, and short-termers.

As you approach your legacy applications in parallel and in conjunction with your business process improvements, consider mapping out your systems into four quadrants--defined by strategic or operational along the Y axis and nondifferentiating or differentiating along the X axis. The delineation of strategic vs. operational should be clear with strategic defining the mission-critical or company purpose systems--some combination of admin, sales support, commissions, and claims. Operational defines the support systems with typical examples including accounting, payroll, reserving, and reinsurance.

Differentiation refers to the degree to which a system provides a market advantage, or a competitive edge, that is unique and sustainable. Sustainability is a key criterion that requires careful thought, as short-term advantages are quickly replicated, making it hard to recover the investment costs. Keep in mind there are situations where an operational system could be differentiating as in the case of a uniquely designed reserving or pricing system that taps into an experience base not available to competitors. And there are cases where a strategic system may not be differentiating, with possible examples being life claims or billing.

Where each system falls is very company-dependent because there is no vanilla categorization; a company might have leveraged its resources and intellectual capital in innovative, unique ways such that systems land in unexpected quadrants. Using this approach, there may be mission-critical systems that are nondifferentiating and that therefore would be subject to a different solution than those that are differentiating.

Once you have categorized all of your application systems into these quadrants, determine the total costs of these four possible options for each:

o Keep the current version current. Do this only if uniquely applied modifications differentiate the company in the marketplace and the investment in staying current can be considered strategic, not simply maintenance. This option also assumes the product complexity is such that continuous regulatory and performance changes make applying the new versions a necessity. Because it is a differentiating system, having a pool of external, vendor-supplied resources (costly though they may be) ensures if the need arises, there are adequate options available to make the necessary changes.

o Freeze at the current version level. This approach should result in the discontinuation of maintenance fees, which would generate the savings necessary to fund the environmental and industry upgrades that would otherwise come from the vendor. This approach assumes, again, a heavily modified system that is coupled to a much less dynamic product. In this case, there may be no strategic differentiation involved; if the system is nondifferentiating, this approach is valid only if the internal resource drain is estimated to be marginal. If it is a differentiating system, the cost is a strategic investment and part of doing business, but serious consideration nonetheless should be given to replacing or porting to a competitive, current technology platform, such as Web-enabled servers.

o Transition to a new, current technology system. For core applications such as administration, commissions, and client management, the transition to a Web-enabled, preferably server-based application suite would be wise. This positions the company for long-term strategic enhancements, simplifies the incorporation of features such as STP, accelerates speed to market, and shifts the TCO from internal maintenance staff to a split between vendor licensing fees and strategic development staff. For any mission-critical, vendor-based system--regardless of the degree of modification--this is the preferred path.

o Outsource. This can be a derivative of either option 1 or 2. As a derivative of option 1, outsource as a service to the vendor that licenses the package. It will price out the cost of keeping the modified version current using its staff--more costly, more expertise, less time, net difference--while the company frees up strategic internal resources that can be devoted to market-differentiating systems and functionality. The option 2 derivatives is the freeze-and-offshore one, which basically entails using lower-cost resources to replace the vendor-provided feature support and releasing the internal staff time. This option should be considered for nondifferentiating systems.

Partnering for competitive advantage, business-enabling, and incremental improvements has become the new mandate. Any dreams or nightmares (depending upon your point of reference) involving multiyear investments in elegant technological solutions have given way to tangible, staged implementations of meaningful and measurable improvements that deliver effectiveness, efficiency, flexibility, and competitive advantage.

For the strategic enterprise, the long-sought alliance between technology, business, and the intellectual capital of the work force finally has come to fruition. Make sure your company is among those that will benefit from this alliance over the coming years.