Continue in 2 seconds

The DSS Mantra

Published
  • December 01 1998, 1:00am EST

"Give me all the data I want, but just the data I need, and only when I need it." -The DSS Mantra> During the past year, I've talked with many companies that have deployed decision support tools or are in the process of doing so. Two things have become clear.

One, most users ­ typically 80 percent in any company ­ have lightweight information requirements. Two, these users are looking for a decision support environment that is best summed up by the DSS mantra quoted above.

The 80 percent of users with lightweight requirements are called "information consumers" because they typically consume reports or information objects developed by others. They are also known as "casual users" because they access the warehouse on an irregular basis, usually to review a particular report or update a budget.

Whatever the label, these users typically want less in a decision support tool than we provide them. Yes, less! In decision support, less is more when dealing with the casual user crowd.

Unfortunately, most companies provide casual users with decision support tools designed for power users, and this leads to potentially dire consequences. Outfitted with the wrong tool, casual users will minimize their use of the data warehouse, causing executives to question the wisdom and value of their data warehousing investments. This often leads to abandoned or stalled projects.

To avoid putting your data warehouse project in jeopardy, heed the DSS mantra. Use it as a guide to help you make decisions about how to architect a decision support environment and select tools for end-user access and analysis. Let's examine the mantra step by step.

"Give Me All the Data I Want..."

Your casual users are knowledge workers whose jobs depend upon information. They may be executives or managers, customers or suppliers, or even front-line workers. They want to be able to access any information in the data warehouse given reasonable security parameters. In essence, casual users are saying, "You've built a data warehouse, now don't shut me out. When the situation calls for it, I want to be able to access any piece of data I want."

It's highly unlikely that these users will ever want or need to access the entire data warehouse, but that's not the point. These users are like people who buy sports utility vehicles, but only drive on highways. They want the assurance of knowing that they can drive off road if they ever need or want to!

"But Just the Data I Need..."

At the same time, casual users don't want to be overwhelmed with information because that makes their jobs harder. They don't want to spend inordinate amounts of time sifting through reams of data to get answers to their questions.

Casual users don't want to explore all the data in the warehouse, they simply want to review a consistent set of data on a regular basis. They only want to drill down and explore the data when they see something awry ­ an exception to the rule in their report. At this point, most casual users will probably ask a business analyst to do the "digging" for them.

Thus, what casual users really want is a parameterized report ­ a report that only delivers the information they want to review on a regular basis. Ideally, this report also comes with a cube of data that gets exposed only if the user wants to drill in on an exception item. Unfortunately, most of the tools we deliver to users prompt or force them to explore the data, something they are not inclined to do.

"And Only When I Need It."

Here's the tough part. Most casual users want to spend as little time as possible analyzing data. They don't get paid to analyze data; they get paid to make decisions and take action.

Thus, casual users would rather not have to log on to their computer, open up a decision support program, find the repository of reports, scan that repository to find the right report, open the report, scan the data and discern any interesting patterns or trends. This is too time-consuming and complex.

Casual users would rather be notified if there's something in the data they should look at. They want to be alerted via e-mail, pager, fax or the Web when there's an exception item they care about. Better yet, they want the report ­ or just the data in question ­ delivered to them via one of these channels, so they don't have to go through the process of accessing the report as described earlier. Going a step further, users want the system to recommend action based on predefined business rules.

This personalized delivery of exception reports requires new sets of technologies that are just coming to market. It embodies triggers, alerts and user-defined agents, as well as a publish-and-subscribe architecture for accessing, manipulating and storing any type of information object.

Select the Tool that Fits

Selecting a decision support tool is like buying a pair of shoes: you need to select a tool that fits your users. This rule is obvious, but it's easy to violate.

For one, most commercially available decision support tools are primarily geared for power users, and these users tend to have undue influence in the selection process. Consequently, companies are more likely to buy tools that cater to the needs of power users than casual users.

Second, it's hard for us who work with data in a technical environment to understand how complex it is for others to analyze data with decision support tools. Most are familiar with reviewing paper reports, and don't have the time, interest or capacity to learn new ways of analyzing that data. Although to us, query and OLAP tools provide the ideal environment for analyzing data, most users find these tools confusing, complex and intimidating.

So remember, the DSS mantra. Repeat it frequently. Let it guide you and your casual users to decision support nirvana.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access