Changing Information Technology (Part 1)

Yesterday's Technology: A Brief Scenario

The year is 1989. Ed, the risk manager, receives a call from Mary, the operations manager, questioning her division's allocation for product liability insurance. She asks to see claim history to justify the allocation.

Fortunately for Ed, his department has access to a Risk Management Information System (RMIS), the state-of-the-art tool for managing insurance information. Ed summons his data coordinator and explains the information needed. Ed cannot order a report himself because users of the RMIS must have a unique "ID" and Ed has not purchased one. Ed did not want to spend the money or learn to navigate through the system and write the complex code needed to generate reports.

Ed's data coordinator writes the report and puts it in the queue for overnight processing. Mary is anxious to see the data, but like everyone else, she knows it takes time to get data from a computer.

The next day Ed's data coordinator brings in the report. When Ed reviews the report he sees that there are many more claims in the prior year than he remembers and the cost is much higher. Ed knows he cannot give the data to Mary like this because she will ask what happened in the prior fiscal year and he cannot give her an answer. Ed summons his data coordinator to develop a follow-up report. If his coordinator fully understands his request, and writes the correct code to deliver what is needed, Ed will be one step closer to meeting Mary's information needs by the following day.

The Same Scenario Today

Consider the same situation today. Mary receives her budget package with the liability insurance allocation. Surprised by the amount, she turns to her PC where her online analytical processing tool is still on the screen. She has just used it to check inventory levels and monthly sales figures. She connects to the "insurance" database and chooses to see her liability claim summary. She instantly sees a problem in fiscal 2001. She clicks to have information displayed by product and sees that there are a large number of claims from the discontinued widget line. She remembers that quality problems lead to several injuries during the year and realizes that the cost of those claims is being reflected in her liability allocation. She has her answer and Ed, the risk manager, never knew she had a question.

The scenario above from 1988 is a pretty accurate picture of how information was managed then and how many companies probably still manage it today. The scenario from today may not accurately depict how any single company is managing its information, but what it does depict is how a company could manage its information if it took full advantage of existing technology. Few companies do because few can change business practices fast enough to keep pace with the breathtaking changes in technology.

How Technology is Changing the Way Information is Managed

The insurance business is perhaps as dependent on information as any other business. Information about prior history is used to set prices, evaluate profitability, and to shape future management strategies. It follows that changes in information technology will have a major impact on the insurance business. To predict how the insurance industry might change requires an understanding of what has changed in information technology.

Essentially, there are six technological advancements that are driving changes in business practices. They are listed in Exhibit 1 and described in detail below.

Hardware Development. The Internet is often credited with being the driver of the recent information revolution, but hardware development has been as much a part of the revolution if not more. The cost of computer hardware has been dropping continuously for several years even as computing power and the ability to store data has improved significantly. Today, a business can manage data on a $6,000 server that required a six-figure mainframe computer just a few years ago.

Perhaps the greatest significance of recent hardware improvements is that equipment cost is no longer a major barrier to application development and hosting. For example, a few years ago there were few organizations with the ability to host a claim database. They included insurance companies, third-party claims administrators, and a few specialized RMIS vendors. Today, most organizations have the hardware needed to operate a claim database and could elect to develop one if they were dissatisfied with the price or utility of systems offered by the traditional vendors.

Data Portability. It is no secret that the Internet has made it easy to move large amounts of data quickly and inexpensively. It has also made it possible for computer systems to communicate with one another more easily. The differences this leads to are significant. First, it is now possible for users to "fetch" data from a remote system when they need it rather than to rely on the host system to "push" the data to the user on a fixed schedule. It is also much easier to transmit "data" rather than reports. In other words, a user may receive a spreadsheet with data rather than a paper copy of a report by fax or mail.

Another advancement is that systems can now draw information directly from other systems rather than relying on a user to draw data from one system and reenter it into another. For instance, to report a workers compensation claim, an insured might need to draw information about the injured employee from its own human resource system and then relay the information to an insurance company operator who would enter the data onto the insurer's computer system. Today it is possible to eliminate the middle steps by having the insurer's system contact the employer's computer and draw the information directly. Not only is it possible, but it is already an established practice for some employers and their insurers.

Virtually Unlimited Access. Web applications also offer virtually unlimited access. Before the Web, the only way to access a remote system was through a dial-up connection. The system provider had to install phone lines and hardware to support the maximum number of anticipated concurrent users. The cost of this had to be amortized over the total number of users.

Web applications offer a "stateless connection." Each user is only "connected" for a fraction of a second as commands pass through a Web server. This means that large numbers of users can have concurrent sessions on a Web application without the need for one phone line or one Internet connection for each user.

The significance of this is that there is little marginal cost to add a new user. This is another factor that is changing the economics of the information business.

Flexible Security. The processing speed of today's computer hardware allows Web applications to include very complex security commands. Web applications work fast enough that every request submitted by a user can be compared to a security table to determine if the user is authorized to perform a command or to see the information being requested. This means that a single Web application can be designed to meet the needs of a very diverse group of users. For example, an application designer might anticipate that both an accountant and an occupational health nurse would need to view data managed by an application. The accountant may need to see the status of premium bills or the total cost of claims falling within a self-insured retention. The occupational health nurse would need to see adjuster diary notes on the diagnosis and treatment recommendations for an injured worker. The problem is the accountant has no need to see file notes on medical issues and the occupational health nurse has no need to see financial data. And, more importantly, each may not be allowed to see the data the other needs to see.

Even though the accountant and the nurse do not need to see the same data, the data they both need likely resides on the same system. For instance, the claim reserves the accountant is interested in seeing are contained in the claim files that hold the notes the nurse needs to see.

A Web application can readily manage a complicated user access situation such as this. The designer would simply create a "role" for both types of users. The role of every user would be defined in the user access or security table. Each page of the application would contain logic that determined which types of users could see the page—or even parts of the page—and execute commands. For instance, the accountant could call up the "Claim Summary Page" and see a link to "Financial Data," but no link to "Adjuster Diary." The nurse could call up the same page and see the "Adjuster Diary" link, but not the "Financial Data" link.

The alternative to building an application with flexible security would be to build a separate application for every class of users. This would be much more expensive and would probably results in the needs of some users not being met.

Design and Maintenance Advantages. New Web authoring software is also making development faster and easier, which leads to lower cost. New authoring tools will provide prebuilt utilities for common functions that can be incorporated into applications rather than built from the ground up. A good example of this is the simple Web counter that registers the number of visitors that access a Web page. The first person to use a Web needed to build one. Today, server operating systems provide detailed Web usage reports and a programmer does not need to build a customized counter to obtain reports on site usage.

System maintenance is less expensive because users of Web-based systems typically need no software other than a Web browser. Web browsers are free and virtually everybody who has a computer already has one and knows how to use it. Because there is no software on user machines, there is no cost to install it or to issue upgrades when products are updated. If new generations of the application are released, the owner of the Web application merely updates the software on its system and the new product is instantly available to all users anywhere in the world.

Finally, if Web development has a cost disadvantage it is that technology is advancing so fast that Web application development is continuous. In the past, a periodic system overhaul may have been adequate, but Web applications must be continuously improved and upgraded or risk becoming obsolete. This means that a system owner must make continuous investments in its systems.

Intuitive Software. The "intuitive" nature of Web software is another advantage because it reduces training and support cost. Intuitive means that instructions are in plain English and commands are given by pointing and clicking. Users do not write code and do not use function keys to navigate through the application.

Advanced Reporting Tools. Presumably, one of the greatest opportunities that arises from storing large amounts of exposure data in one place is the ability to analyze it for trends so that future loss prevention efforts can be shaped by analyzing loss history. Analysis is usually done by writing reports that find, sort, and display selected data.

Perhaps the greatest differentiation between traditional mainframe RIMIS systems and Web-based systems is the speed and flexibility of report writing. This was illustrated in the opening paragraphs of this article.

There are two types of report writers that are commonly used with Web applications. One is an "ad hoc" reporting tool that allows a user to build custom reports by selecting the data fields to be displayed and the format in which they are to be displayed. The other type of tool is an online analytical processor (OLAP). An OLAP is used to perform analysis at the summary level. It simplifies the production of summary reporting but does not provide the same flexibility of an ad hoc report writer.

To illustrate the difference, an ad hoc report writer would be used to generate a list of all auto claims in litigation where the insured's driver was at fault and the expected cost is more than $5,000. An OLAP tool would be best to analyze year-to-year workers compensation performance and to identify the major causes of loss.

OLAP tools are very simple to use whereas ad hoc report writers may require a strong knowledge of the software and of the underlying databases. Because OLAP tools are more "user-friendly," they may eventually become a standard tool on the typical business desktop. (See Exhibit 2 which describes OLAP tools and the future of these tools.)

As an alternative to a report writer, an application designer could try to anticipate the types of reports users might need and build them into the business logic of the application. Then users could generate the reports whenever they were required. This is generally a poor solution for report writing needs because it is expensive to write reports in this fashion and it limits the user's ability to analyze data.