As competition ratchets up in the insurance industry, insurers need to hone their operations by using the power of information, but the industry is still struggling to overcome the limitations of the early generations of insurance systems.
Core legacy systems were designed to process transactions--deliver quotes, produce policies, collect premiums, pay claims, manage compliance and handle accounting. Essentially, they processed inbound and outbound cash. They weren't built to transform data into actionable information for decision-making or operational management purposes.
As a result, most insurance companies are sitting on a potential goldmine of information they can't get to--at least not without entering into a multiyear data project costing many millions of dollars.
However, this is changing as products that let insurance managers take the pulse of the business are becoming commercially available.
Today, market cycles have been compressed down from six or seven years to one or two years. The new generation of systems must both handle transaction processing and become true information and "business intelligent" systems.
New technology that enables this is based on fast, inexpensive hardware and industry standard software building blocks. Low-end servers that now hold terabytes of data can be assembled quickly to create more robust, flexible storage systems than is possible with mainframes.
On the software side, some major software companies now offer powerful off-the-shelf components that solution providers can use to create more robust, powerful systems than were possible before. It's like using prefabricated panels to build a house instead of pounding in every nail by hand.
This speeds development and provides more robust capabilities, allowing developers to focus on the business application at hand.
A powerful reporting and business intelligence system requires the following three main components.
o Data warehouse and data marts.
The first component of the new breed of data warehouse and analytical reporting solutions is the data warehouse itself, because trying to tap application databases directly would adversely affect those systems' performance.
The data warehouse defines the underlying insurance data model with all of the business objects within a carrier organization and defines the logical and physical relationships among those business entities. Extract/transform/load scripts (ETL) are used to extract data from the core application databases and load the data warehouse.
In the past, carriers have typically had to write their own ETL scripts to extract data from multiple databases, transform that data to meet business needs, and load it into the warehouse.
However, newer solutions now offer pre-written ETL scripts that shortcut the process of loading the warehouse. This dramatically reduces project schedules and associated costs and allows for faster implementation of the solution.
In a similar fashion, data marts that provide pre-defined reporting views of information are pre-loaded with data from the data warehouse. This is another area where custom work was required in the past, but no longer.
Because of the complexity of the interfaces with the underlying applications, as well as upstream and downstream systems, most reporting systems aren't "plug-and-play." They required large-scale projects to implement.
But instead of this type of massive project, newer products allow carriers to implement the solution in less than a year and at a significantly reduced cost.
Additionally, data warehouse solutions employing standardized interfaces and Web services can interface with any existing insurance system, whether old or new.
o Report authoring, publication and distribution.
This is the reporting user interface that enables users to define and publish reports to the various groups of users--whether they are internal or external users. This facility allows the user to manage report listings, user subscriptions and notification lists that specify which users will see which reports, based on security that is segmented down to particular data elements.
The solution can notify users automatically and can distribute reports by e-mail and, if desired, to PDAs, cell phones and other mobile devices.
o Business intelligence and advanced analytics.
Business intelligence and advanced analytic capabilities set data warehousing and reporting solutions apart. Without them, all that's delivered is a reporting solution.
The new breed of data warehousing and reporting solutions have these capabilities built-in or "out of the box." Not only is online analytical processing included to crunch the data and identify trends, but more important, so are the definition and management of business scorecards and key performance indicators that transform traditional reporting systems into operational intelligence solutions.
These capabilities act as sentinels that alert the carrier when key business thresholds are being approached so that the carrier can take action before the issue adversely affects its business.
This allows the carrier to continually balance its risk portfolio, aggressively manage its distribution partners, and measure supply-chain partners to a degree previously not possible.
With all three components, you can build a system that goes beyond reporting and becomes a key asset in running the business.
With traditional reporting systems, batch reports are run monthly or quarterly. By the time you've received a report and analyzed it, something may have gone terribly wrong or off the tracks, and you'll have to scramble to prevent more damage.
Wouldn't it be much better if you could easily identify dangerous trends and budding opportunities early?
Now you can. A reporting system with business intelligence lets managers create scorecards and set warning indicators.
For instance, let's say that you want to limit your exposure in a certain geographic area, or line of business or product. Now you can have the system automatically alert you when you're approaching the threshold so you can take action.
Let's say you've reached your limit in Florida. You turn off the spigot.
You can set up virtually any alert you want, as long as the raw data has been captured in one of the core systems. Your indicators might include investment returns or the number of days to adjudicate a workers' compensation claim for, say, temporary total disability.
The alerts can be set up to be triggered when either good or bad things happen. For instance, let's say you have a program to insure municipal vehicles. The system alerts you that losses are very low. You can now open up the underwriting floodgates, write more of it and reset a higher threshold.
Business intelligence that converts data into information you can act on is a strategic advantage for any insurance company. When you know what's going on with your business at all times and don't have to guess, you're in position to win.
