Printer Friendly

Making sense out of data.

Grocers are blessed (or cursed, depending on your point of view) with having access to mountains of data generated in the course of doing business. However, as has often been pointed out, they don't always use that data to their advantage. That may change, though, thanks to the emergence of executive information systems (EIS), which are designed to grapple with and make sense of large amounts of data.

Among technologies being tested by supermarket retailers and wholesalers, EIS packages can be fairly described as "hot." In FMI's Information Systems Study for 1994, 13% of 57 companies surveyed said they were in "advanced use" of EIS, and 27% claimed to be in limited use. But the most telling statistic was that 22% said EIS was "being developed" - the highest percentage in that category for any technological application (except for POS EAN scanning, which also scored 22%).

EIS systems collect a lot of information and look for the exceptions - where the problems are, says Alan Levin, president of Decision Support Technology, a Cambridge, Mass.-based EIS consulting firm. Some EIS packages use color codes - red and yellow - to flag problems. EIS "gets you in front of problems to see them as they're happening so you can respond quickly," says Levin. Adds a data administrator for a Western wholesaler, "It's getting so complicated that you can't survive unless you have tools like this." (See sidebar for a further definition of EIS.)

EIS has evolved from its original role a decade ago as a tool for high-ranking executives "with corner offices," says Eric Schnadig, a Washington, D.C.-based marketing specialist for Kenan Technologies, Cambridge, Mass. "EIS has evolved into something that can be used by everyone, which is now what the `E' really stands for."

At Hannaford Bros., Scarborough, Maine, regular users of Hannaford's EIS system include 30 to 40 people, ranging from financial analysts to district managers. (Store managers don't have access to it.) Moreover, says Al Carville, Hannaford's vice president, information systems/technology, "everyone looks at the same information as soon as it is available."

Users of EIS systems need a minimum amount of computer literacy, such as familiarity with Windows desktop systems. Hardware requirements include a network of Windows-based PCs connected by a local area network (LAN), with access to a Unix server with a storage capacity of 10 to 20 gigabytes. Hannaford's system runs on 386 or 486 PCs, while its database resides on a mainframe. Data from store level can be transmitted to headquarters by means of a wide area network (WAN) or sent on tapes through the mail, says Schnadig. In Hannaford's case, data is transmitted via a satellite network.

Given the often substantial expense of EIS packages, Schnadig suggests that EIS users take a "targeted approach" at first. "Identify your needs first," he says. "Be careful not to run in too many directions. Ask yourself, `Am I ready for this?'" The "entry cost" for Kenan systems is $25,000 for a five-user client-server configuration, which includes a database and applications. But Levin puts the total investment for an EIS system at between $150,000 and $500,000.

Hannaford has used its EIS system (from Comshare Retail, Wilmington, Del.) for six years. Carville describes the EIS package as a "fairly effective" performance measurement tool that "provides senior management with current information on sales, payroll and financial statements in a very easy-to-use graphical format." EIS systems, he adds:

* Have drill-down capabilities, so that users can "come in at the district level and click to get down to stores."

* Can calculate measures like average sale per customer, or average transactions per store per week and present the results in graphical form.

* Show the variance from budget (exceptions) for each performance measurement.

* Allow users to easily make ad hoc inquiries into things like "bad-check write-offs for all stores."

The database administrator for the Western wholesaler, who requested anonymity, has been working with an EIS system (marketed by Kenan Technologies) for about a year. The EIS package resides on a "little desktop" PC running Windows NT. The database, which has a 1.5-gigabyte capacity, holds up to two years of weekly sales data and up to eight years of monthly, quarterly and yearly data. The data that drive the EIS system are derived from invoices for retail purchases. However, the data administrator says he plans to eventually use scan data instead. Thus far, the users of EIS have been high-level executives, but the next step is to provide access to corporate-level people in warehousing, accounting or retail support.

The data administrator pronounces the EIS system a major improvement over the "query" software previously used by the wholesaler in analyzing its business. "It took quite a bit of smarts to know how to pull data together when using a generic query product," he says. In addition, it took "tremendous resources to crunch through the data and generate [reports]. The sheer time involved was prohibitive."

The EIS package allows users to more efficiently:

* Compare sales during different time periods for individual stores or for the wholesaler as a whole.

* Calculate margins for certain departments or types of retailers.

* Generate state-required tobacco sales reports.

* Calculate 13-week averages for items stored in the warehouse. "Hot" items are reported to stores.

* Assess the performance of private label vs. national brands. "If someone is falling off in a private label category, he can ask, `What's the problem? Am I too high priced?'" the administrator says.

Items can be grouped in nine ways, such as single products, categories or departments. Retailers are grouped 10 ways, such as by state or business affiliation (such as IGA). Time spans include week, month, quarter and year. The wholesaler has some requirements that the Kenan package (but not some competing packages) is able to meet, such as a week that runs from Friday to Thursday and a fiscal year that does not coincide with the calendar year.

The data administrator explains that in the past, if the wholesaler wanted to know how much of a single product was purchased by a particular independent retailer in a single week, someone would have to go through every invoice associated with that grocer and add up the sales for each one. The EIS system, however, does that sort of addition automatically and has it ready for a user to access, he says. "It makes reporting of summarized data very fast."

Knowing each retail customer's sales for a given product line, the data administrator can then compare each store with all other stores served by the wholesaler. "That tells us how significant a retailer is to us," he says.

By getting trend numbers on various store departments, the wholesaler can then be in a position to take corrective action. "It helps us to home in on where we need to improve," he says. "It may mean we should work at getting better deals or concentrate better on certain ads."

The biggest challenge in implementing the EIS package, says the data administrator, was "getting clean data to load into the system." Coming from different sources, the data was not all formatted the same way, and as such had to be put into a uniform format before it could be used in the EIS package's database. (Otherwise, you have a GIGO situation: garbage in equals garbage out.)

Overall, he says, the EIS package is "a better way to track and calculate things. I can't imagine going back to the old way." Under the old system, he was dealing with too many unrelated files; one might hold year-to-date figures, another would hold month to date. "You don't have to think about which files hold which data," he says. He says that just for internal use, the payback is "real fast."

Park City Group, Park City, Utah, has an EIS software product called Score Tracker, which the company says is being used by Winn-Dixie and Bi-Lo. It gives daily exception and drill-down reports, pointing out trends and problems in such variables as sales and labor for the company, store or department level. For example, it tells whether a particular department met its budgetary requirements.

In its latest products, Park City Group takes EIS beyond the level of reporting strategic information by making "realtime suggestions that can influence outcomes," says Randy Fields, CEO of Park City. This system, called "Action Board," was released this past January and may be installed by Winn-Dixie this fall, says Fields.

Action Board also works across all levels of operation. On the departmental level, for example, it might alert the bakery manager that he could run out of muffins later that day and should therefore make 52 more muffins right away. Or it might tell a store manager that he needs to open more checkout lanes to accommodate the business that day and should adjust his labor schedule accordingly.

On the district manager level, it might inform the district manager that a store is running significantly behind in sales and should be visited right away - rather than at the end of the month, as would typically happen. Both Action Board and Score Tracker cost around $200 to $400 per store - among the least expensive packages around.

What is EIS?

EIS (executive informations systems), sometimes called decision support systems, refers to systems that convert data, which by itself is meaningless and useless, into information, which has both meaning and usefulnesss. "EIS gives you detailed analysis that goes beyond trends to how the business is doing today and how it will look in the future," says Eric Schnadig, marketing specialist, Kenan Technologies, Cambridge, Mass.

Here's another way of looking at EIS: Over the past 20 years, retailers have focused on what is called on-line transaction processing (OLTP), the process of gathering data, whether it be at the checkout (POS scan data) or in accounting, with, say, payroll data. Now there is growing interest in another approach, called on-line analytical processing (OLAP), which takes the accumulated data and fashions it into something revealing about the business. For example, OLAP systems may take gross margins and organize them by category, time period and group of stores, and compare that to comparable figures from the preceding year.

Kenan's EIS package incorporates what Schnadig calls multi-dimensional database technology, which is a fancy way of saying that data can be analyzed from "a number of different perspectives." For example, sales can be tracked not only by product, but also by time, location or day of the week.

EIS systems provide analysis that can form the basis for corrective action. A graph that points out a troublesome trend can be sent by E-mail to store managers, who can attempt to reverse the trend. If gross margin is down for a particular store, corporate executives can suggest that the store manager call store XYZ, which achieved a good gross margin for that product. The following week, the margin is reported to see whether the corrective action worked.
COPYRIGHT 1995 Stagnito Media
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:includes related article on executive information systems
Author:Garry, Michael
Publication:Progressive Grocer
Date:Jun 1, 1995
Previous Article:Where the customer is king.
Next Article:Checkout cash cow beefs up.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters