DECIDING FACTORS

In the realm of management information systems, few issues loom larger today than the ability to provide information to all levels of an organization for making better business decisions.in determining which companies emerge as industry leaders heading into the 21st century.According to retailers and wholesalers, many of the technological capabilities now exist or will soon emerge to develop comprehensive

In the realm of management information systems, few issues loom larger today than the ability to provide information to all levels of an organization for making better business decisions.

in determining which companies emerge as industry leaders heading into the 21st century.

According to retailers and wholesalers, many of the technological capabilities now exist or will soon emerge to develop comprehensive decision-support and EIS solutions. However, a wide range of issues, from training nontechnical staff to changing corporate culture, still must be addressed before the systems' full potential is realized.

The desire of MIS executives to delve into the enormous power of decision-support and EIS is clear. What is less clear is how soon the industry will be ready to take advantage of these tools. Also at stake is whether all industry players, or only the most innovative companies, will make the investment and forge ahead in this area. Here is how SN's discussion with five industry leaders went on the subject:

SN: How big is the need for decision support and EIS?

BILL MAY: The use of decision-support systems is becoming increasingly important as the daily data we have access to multiplies exponentially. Decision-support systems are one of the hot trends not just in our industry but in many industries.

Because of the availability of data, decision-support systems are going to become a cornerstone at all levels of the organization. Employees are going to have to be able to query data within whatever data bases they need to and be able to do it on an ad hoc basis, real-time, without technical intervention. That is the issue.

RAY HAMILTON: Decision-support systems are the reason for our existence at any successful company. The day when executives, and I mean at the CEO level, don't have PCs at their desks is rapidly vanishing. The idea that an executive can have information to make strategic decisions or even tactical decisions has a lot of merit.

DAVID HAYES: Decision-support systems are crucial for direct advertising or niche marketing because of the tremendous amount of data we are gleaning from the store level. It is important that we be able to drill down into that data to make marketing decisions.

PATRICK STEELE: One thing constant in our business is change, and the need to be responsive to change is going to become more and more critical as an issue. Companies that respond more quickly to change are going to have a competitive advantage, and decision-support and EIS can satisfy that need. Successfully managed businesses must have timely information to respond to changing trends and conditions as quickly as possible.

If companies can make use of the information they have, if they can create a large data warehouse and decision-support system that their people can tap into to analyze trends, they can make better decisions. They can make sure their stores are merchandised properly, carry the right products and have the right prices. They can gain an edge on serving their customers instead of competing under the old rules and running their stores like everyone else.

It's a very hot topic. These systems can provide a true analysis of what is happening. They fit in with the whole model of ECR and having the right items on the shelf, getting the best procurement costs possible and delivering product to the store as inexpensively as possible.

SN: Are companies beginning to take full advantage of these types of systems?

HAYES: I don't think EIS has advanced very far yet. We are still probably behind where we should be. For one thing, data is not always 100% accurate. It may be accurate when it comes to looking at trends, but it is not numerically accurate. Everyone is in the same boat. We're also all experimenting with data warehouses to hold the data.

MAY: We are a good two, three, four years away from the point where people at all levels of the organization sitting at their desks will be able to access information through a decision-support system. It is going to take time just to get these systems on everyone's desk and provide training on how to use them.

STEELE: Decision-support systems are just starting to move to the forefront of industry attention because of the technology now available. It's much easier today to store vast amounts of data and use it for analysis and spotting trends. Companies are going to have to pay close attention to this area in the next couple of years. These systems are going to help drive the direction of how each of us responds to customer needs.

But the technology is there to make this happen from both the data warehousing and applications standpoints. There has already been some pretty good success using these tools for managing categories in our industry.

PETER ROLANDELLI: When we talk about EIS, we're talking about a work force that is not yet skilled enough in the basic technology tools to go through and analyze these programs. They simply have not had the time to makes themselves proficient in using these types of analytical tools.

What we can do today is develop simpler versions of EIS that allow executives to look at some key numbers. We can find out what an executive considers the key numbers and then automate the Monday morning reporting package to provide that data. But when we get down to decision-making, there still has to be someone who can delve deeper into the analytical program. It is our responsibility to get all the information we're going to need and do so economically. That's the key. MAY: EIS has not caught on yet, but as the work force becomes more computer-literate, executives will jump into that technology. It's crucial to make sure that the people using EIS have a high degree of comfort with it. They have to believe that the information is accurate and they have to know how to make use of the data. There's a long way to go in this area. EIS can play an important role, but the programs have to be carefully constructed. There has to be training and there has to be dedicated support staff. I don't see any company really using these systems in a full-blown way yet. But they will become increasingly important over the next few years. We will get there. It's one of the strategies we have.

SN: In which areas do you see decision-support systems and EIS playing an especially crucial role?

MAY: I think these systems will be especially important in the procurement and operations areas. A warehouse manager needs to be able to go on line and see real-time what is taking place, how efficiently cases moved yesterday and what labor costs per hour were during the past 24 hours. In procurement, a manager must be able to pull up a report on a back-to-school sale last year, look at how each SKU did in terms of movement, percentage increases from a year earlier, and offer a recommendation on how much to prebook this year.

I also see it playing a major role in the financial area by analyzing precisely and quickly what it costs us to work with a vendor or take part in a program. What is the cost of business? So I think it is going to have applications throughout the organization. Eventually, people will be able to call up and design reports on the fly, immediately, to make informed business decisions. STEELE: Decision-support systems take place more at the item level for analyzing the strategic impact and potential results of how a category will perform if certain changes are made. A big area is merchandising, such as category management, data-base management and data-base marketing. EIS, on the other hand, will be used for determining company or department trends at a higher level, often as the top layer of a category management program, for instance.

HAYES: There's no limit to the types of information that could be generated as the data becomes more accurate and is placed in the hands of the end users. The trick is designing the data base. That requires a tremendous amount of planning.

SN: Could you address some of the challenges involved in taking advantage of these systems?

ROLANDELLI: One challenge is that potential end users don't yet know what they want from these systems. Definition is a big problem here. How do I know what pieces of my company's operations will support the decision if I haven't figured out what the decision is that has to be made? Will it be a decision to build a new store in a specific area? If so, we can pull a lot of complicated issues together and focus on that. But if the decision is how to increase sales in the New Orleans division, that's a tremendously complicated and much different problem.

Not only do we have to determine what the question is, but we have to figure out all the pieces involved in making that decision. HAMILTON: One critical issue is remote accessibility, because most executives travel three or four days a week. If executives can log on from their laptop computer and access key numbers, sales, labor numbers, they can act on information from that day.

The demand for remote accessibility leads, in turn, to the need for a wide-area network as a solution for desktop or portable PCs. There has to be a way to get the software programs and the information needed displayed on the screen. These systems must be capable of turning data into information that can be used to make decisions.

MAY: I think a big difficulty is the sheer number of software programs out there today that promise to deliver every possible application and some we never dreamed of. But on closer examination, it often becomes more clear that many of these tools are being oversold. Eventually, there is going to have to be a shakeout to where a few legitimate tools emerge that end users, nontechnical people, can use.

Many of the tools today are cumbersome and require technical intervention. The systems are not there yet in terms of that flexibility. We have many requests, for example, from people who want to query a data base about a particular customer or strategy and need the information that afternoon. I can't do that today. Those requests still require MIS intervention to set up parameters and drive the software. That is a huge problem. HAYES: As with any application development, it is critical to spend a lot of time with the end users and find out what kinds of queries they may want to make. We should all start out slow in this area, build some successes and then add to those successes. The key to this is design.