Manage-Wise

Decision-making: more than a spreadsheet

We learn from our mistakes. In the mid-1990s, we opened a store in Amarillo, Texas. At the time, the store offered the best and most customer-friendly technology in the supermarket industry. The bakery department was an area of particular focus for our company. We had high expectations, and we thought the department would serve as a difference maker in terms of freshness and quality—important considerations for all guests visiting the store.

Like most retailers, we based our decision to build the store largely on data supplied by real estate sources detailing growth trends, the potential available dollars in the trade area, the per-capita weekly expenditure for food, and the simple attraction the store represented. In those days, we had an undeveloped marketing department.

In fact, we had three full-time team members and one or two part-time interns working their way through college. The staff dedicated virtually all of its time to advertising, not marketing. As a result, new-store construction was a team decision based on spreadsheets and corresponding data. In the case of our new Amarillo store, the numbers justified the opening.

Each week following the grand opening, our leaders would study the financial reports on the store’s performance. Interestingly, despite great numbers from the store in general, we noticed the bakery department was underperforming. We were not realizing the numbers we had projected. Sitting in our offices 120 miles away and studying the spreadsheets, it appeared we had missed something. Leaders began quizzing the bakery director, who, in turn, began quizzing the bakery manager at the store, seeking answers for the lackluster performance. Initially, the thought surfaced that we might have quality issues, but the recipes were the same as those used in our other stores, and the bakery team had not made changes.

Our attention soon focused on the people in the bakery. Wrongly, we assumed we had to make a change in leadership at the bakery manager position. Sales remained flat. Frustrated and confused by the bakery’s performance, we formed a small team to thoroughly investigate the problem.

We left our offices, made the two-hour drive up the interstate, checked into hotel rooms, and camped at the store for a week. We asked questions of guests and observed the daily traffic. Three days after immersing ourselves in the store, we realized our problem: a large percentage of the guests shopping the store were empty nesters. We had neither a quality issue nor leadership issue. We had a packaging issue.

Empty nesters were not interested in buying a dozen doughnuts, nor were they motivated to purchase anything by the dozen. They wanted smaller portion sizes. We validated our findings by observing similar trends in the meat market, where large packages of ground meat were not selling, either.

Hurriedly, we changed our packaging. Eureka! Sales for doughnuts, bagels and other bakery products increased, and the store’s performance improved.

Since I was serving as marketing director at the time, I felt bad about what had happened. As a result, I pledged to improve my decision-making process. Working with my team, we created a new decision—making model, one that would have prevented the bakery fiasco in our Amarillo store.

The model shown here consists of three buckets of information. One contains information directly from guests, the buyers of our products and services. A second bucket contains information from the sellers—the collective intuition and knowledge of seasoned professionals. The third bucket contains empirical information, data based on actual performance.

Business decisions are challenging enough when leaders possess data from all three buckets, but a decision made using data from only one bucket, regardless of which bucket it is, is problematic. If decision makers hope to bring about satisfactory resolution, they need a balanced protocol. The three-bucket approach to decision- making for organizational leaders is similar to the triangulation method of navigation used by pilots, sailors, and explorers.

Feedback from each bucket allows an organizational leader to move one step closer to confidently establishing a bearing, a position. One data source can provide a faulty signal, but having a second source and, better yet, a third source exponentially narrows the margin for error.

In the case of the first bucket of data, buyers’ intuition is valuable, but only in the context of human emotion. Suppose a supermarket chain initiated a program at store level to stock every single item asked for during the course of a day, a month, or a year. Even with a virtual inventory capability utilizing online suppliers, the space required to manage the physical store inventory would be staggering, and the amount of perishable product discarded because of a lack of sales would make the cost of this initiative prohibitive.

The truth is, buyers can put an organization out of business without even thinking twice about it. A guest who wants one special jar of dressing for a meal she is preparing has little regard for the fact that dressings are sold to the retailer not by the individual unit, but by the case. She is more than willing to buy one jar, but the rest of the case is someone else’s problem. The buyer’s feedback is suspect only because it is born out of emotion rather than reason. The decision to buy or not to buy is one of “feeling”—a decision made from the heart, not from the head. The heart contains telling information, but it is prone to exaggerate the truth and view the world from a myopic perspective.

The second bucket of information contains its pitfalls, as well. A seller’s intuition offers immediate insight, but it too represents a “feeling” decision. Little reasoning goes into formulating a seller’s intuition. Instead, sellers rely on emotional mental imprints rather than on objective reasoning.

For example, supermarkets have a wide variety of selling programs running concurrently. In addition to their own product offerings, supermarkets will allow business partners to stock and sell books, newspapers, magazines, and greeting cards, among other things. A profit-sharing arrangement allows such programs to be mutually beneficial. Full-service programs, where the supply partner ensures adequate stock by having its people service the stores, free up the store’s team members for other tasks.

When shelf space in a store is tight, some store directors may suggest reducing or eliminating such programs. Their intuition tells them they do not sell much of the product because the rack is always full. In other words, products in the store that require handling by the store director and staff are either restocked, which mean they are selling, or dusted, which means they are not selling. Good operators have a mental imprint of what sells and what does not.

These tangible signs of success or failure provide operators with an impression of the movement of one product relative to another; however, with full-service programs, the racks appear to be full all the time by design, and therefore, some store directors may not fully appreciate the movement of the product because they are not restocking the item daily. Their intuition tells them the product is a failure. In short, they are susceptible to being fooled into feeling that a product is not selling when it may be selling just fine.

Excerpt from ‘Built to Serve’ by Dan J Sanders. Reproduced with permission © 2008, Tata McGraw-Hill Publishing Company Limited. Price: Rs 495. Vishwanath_Ghanekar@mcgraw-hill.com