So, let's briefly characterize what is meant by each column:

o Integrated Package is what we traditionally think of as buying a vendor solution or "package." A good example would be a policy administration system that includes rating and print solutions.

o Best-of-Breed Components is the mix and match of assets, often from different vendors, which provide point solutions and integrate readily with other point solutions and also legacy components. A good example would be a rating or underwriting engine or a claims system.

o Framework/Tools refers to the purchase of rules engines, software frameworks, and models to provide a start point to build using powerful and flexible assets. To complicate things further, the new generation of vendor-written core applications looks more like frameworks and tools than packages. This is because these vendors started by either writing or acquiring tools and then used their toolkit to begin to develop and deliver "packaged" functionality.

o "Pure" Build means new original code gets specified, written in either Java or .NET, and tested.

These look like nice neat choices, but of course, in reality, carriers choose combinations of these options to suit their needs. For example, in two recent instances where large, capable carriers decided to build policy administration systems, neither company built a rating engine; they bought one and integrated it. Also, don't expect to find carriers building underwriting or workflow engines or document management solutions. Even the largest IT shops in our industry, which still believe their role in life is to build software, are not that mono-focused. So, caveats aside, what are the pros and cons of each strategy? Table 2 summarizes the main points for consideration with each option.

The options available run from low cost/low risk/low satisfaction on the left to high cost/high risk/high satisfaction on the right. In 2006, my company conducted an industry survey with a group of insurers with the objective of answering the age-old question of virtually every CIO: "What are my peers and competitors doing" with reference to core insurance systems? We managed to create a fairly reliable picture of 23 carriers that ranged from regional to superregional (or about $200 million in gross written premiums to $2 billion). These carriers fell into two roughly equal groups: those replacing core insurance assets and those enhancing and maintaining legacy assets.

Of those carriers that had embarked on a replacement strategy, the results included:

o Only one carrier, the second-largest in the group, was building significant core systems but was integrating with third-party rating, underwriting, and document management solutions. This carrier also purchased most development services from an offshore vendor.

o The majority was using "best-of-breed" solutions to replace core systems.

o Generally speaking, those carriers that were implementing integrated packages had lower expectations their solution would satisfy their requirements than those undertaking other replacement strategies. In several instances, integrated-package carriers felt they probably would change business practices to fit the software rather than attempt to modify the software to fit their business.

o The group in general felt the integrated-package strategy yielded the lowest risk/satisfaction proposition, the build strategy yielded the highest risk/satisfaction proposition, and the difference between the two was far greater on the risk dimension than on the satisfaction dimension.

To take this last point further, there was a definite sense among respondents as to how each of the four strategies rated on both the risk and satisfaction scales. The general sentiment expressed is captured in Table 3. The following points are notable:

o Risk and satisfaction go hand in hand. The more a carrier seeks a custom solution that will provide a good fit with requirements and will be modifiable over time, the more complex, sophisticated, and therefore risky the implementation.

o Risk and satisfaction were not seen as being proportional. While all respondents agreed the greatest potential satisfaction would be created by a software build strategy, most also felt the risks significantly outweighed the benefits.

These findings raise the issue of which strategy or hybrid strategy maximizes satisfaction while minimizing risk. According to Table 3, that desirable combination is most likely to be found somewhere in the framework/best-of-breed space. Let's also remember the point made earlier that the new breed of vendor packages look and act far more like frameworks and toolkits than traditional packages.

There is one other qualifier that needs to be stated here. The subjects of this survey were personal lines and small commercial carriers. The findings stated may be different if the study group represented a different market profile.

The last item to bear in mind is acquisition strategy affects the criteria that a carrier uses to evaluate and compare vendors and their solutions. What we see in a traditional RFP or RFI is a strong focus on functionality and business requirements, and that is appropriate if a carrier's strategy is to purchase and install an integrated package or a best-of-breed point solution. However, this focus is not as appropriate if the strategy is more in the framework/tools space. Here, there is less functionality by definition and more of a focus on software flexibility and the abilities and track record of the vendor in using its tools successfully to build and deploy core insurance solutions.

So Many Choices . . .

This column has been about one specific, early choice a carrier needs to make when undertaking a major systems acquisition. It is a choice, of course, which could be made almost by default for the majority of carriers, especially in the past.

As with so many of these vendor-related issues, it's a case of make a choice or a choice will be made for you, usually by attraction to the "shiniest toy." So, carriers should proceed by making sure they at least are in the right toy store. As the old saying goes: "If you don't stand for something, you will fall for anything."