PROFIT FOUND IN VARIETY, SKU DUPLICATION CUTS

MARC MILLSTEIN

TAMPA, Fla. -- Eliminating excess duplication by slashing stockkeeping units as much as 25% sparked sharply higher sales and profits in three pilot programs.

The programs, at Cub Foods, Bi-Lo and Save Mart Supermarkets, involved reducing product duplication and using the extra space to enhance variety. As a result of the moves, per-SKU net profit and sales shot up by as much as 40%.

The test programs, conducted under the umbrella of the industry's much touted Efficient Consumer Response initiative, were unveiled at the Food Marketing Institute's second annual MarkeTechnics convention here.

In one test at Cub Foods, SKUs within the hot drinks category were chopped by 14%. Rather than reducing sales and profits, though, the category showed considerable gains, said Lynn Olsen, corporate director of shelf management at Cub Foods, Stillwater, Minn.

"We had about a 14% reduction in SKUs, yet product movement, sales and gross and net profits all increased. Two indexes in particular, per-SKU sales and profits, were especially remarkable. They jumped nearly 30% in both cases. The positive results of the test are an indicator that over the long term we are going to gain in store productivity," Olsen said.

The test results covered a 17-week period and involved the coffee, instant coffee, creamer, tea and hot chocolate segments.

"We did a number of things to reduce SKUs, including getting rid of some product size duplication, package type duplication and

flavor duplication, especially in a hot category like tea where a lot of people are entering the market with similar flavors," Olsen said.

"At the same time, when we freed up that space, we added private label, we added new flavor creamers and we added new varieties of tea that we simply didn't have room for before," he said.

Cub Foods also conducted a test in the liquid detergent area. The program focused on reducing size duplication and reallocating the space for better items. The test compared an eight-week post-reset period with an eight-week preset period.

After eliminating a number of lines, the chain "added some private-label and refill lines. Overall we had a 17% SKU reduction, but again, unit movement, sales and profit all went up. In fact, on a per-SKU basis, net profit and sales increases were amazing, jumping close to 40%," Olsen said.

Olsen and other speakers shared the results of their test programs at a workshop titled "How Retailers Are Managing Variety and Reducing Duplication." The tests were based on recommendations from a Food Marketing Institute ECR-related publication released at FMI's annual convention in May. The report, "Variety or Duplication: A Process to Know Where You Stand," was co-authored by Dan Raftery, vice president of Willard Bishop Consulting, Barrington, Ill., who also spoke at the seminar.

"One of the four strategies outlined in the ECR initiative predicted a 1.5% cost savings to the bottom line for this thing they are calling Efficient Store Assortment. Removing duplicate items is one way to make the store assortment more efficient," Raftery said.

"Variety is defined as those items that add to sales, unique items. Duplication is those items that do nothing more than spread sales across themselves because they are not really offering the customer anything new to choose from," he added.

Eric Verhoeven, manager of marketing research at Bi-Lo, Mauldin, S.C., also revealed that a program to reduce duplication yielded positive results.

"The things we learned from the 1993 study were, first, there is plenty of duplication on the shelf. Second, you can reduce SKU count without upsetting your customer. "Third, SKU reduction does not have to have a negative impact on sales, provided you do it right," he said.

The Bi-Lo test involved the diaper category and measured several parameters, including diaper uses, or actual number of diapers, not SKUs, sold; gross and net margins; closure rate, or the number of diaper units sold as a percent of total customer count, and average transaction per diaper purchase.

As part of the study, Bi-Lo delisted 25% of the total SKU count in the diaper category, which translated into pulling 20 items from the shelves and eliminating three brands entirely.

"So what happened? The total number of diapers we sold increased 6.9%, based on a comparison of the first half of 1993 with the second half of 1993. Also, our net margin improved by 1.2%, and while that is only about half-way to the goal we set of 2.3%, it is still progress," Verhoeven said.

In addition, "inventory turns showed dramatic change there. Our inventory turns before the reset were 30; after the reset they were almost 50. "We also saw a healthy in-crease in our diaper market share," he said.

While the closure rate showed no significant change, rising only from 2.5% to 2.6%, the average transaction rose 4.7%, from $7.39 to $7.74, despite some major price decreases during the program, Verhoeven added.

"All in all, it is very obvious that this pilot project was very successful. Not only did it improve our performance in the diaper category, but I also believe that it opened some eyes," he said.

Charlie Swanson, manager of space management at Save Mart Super Markets, Modesto, Calif., also cited positive results from a test program.

"We deleted 24 SKUs that we identified as size duplications, declining subcategories or low movers without unique distinctions. We then reallocated that new space to items we thought would have higher turns," he said.

The results were striking and immediate. "During the first five weeks of the test, product movement was up 20%, sales were up 21%, profits were up 19%, turns were up 55%, and the cost of inventory investment was down by 23%," Swanson said.

"When you first look at these numbers, you would think that we must have been doing a terrible job managing the section before. But like many of you out there, we felt we were doing a good job.

"This program just indicated that with a little more work, we can squeeze a little more out of our shelf space and do a better job. These were very positive results in every area. "But they do reflect only the first five weeks of our test period and will likely smooth out to some degree by the end of the full test," he said.