Procter & Gamble, on the verge of acquiring Gillette for an estimated $57 billion and extending its product footprint, is behind the scenes making what might be even more striking moves that could shake up the industry.

P&G is developing and deploying technologies and revised business processes to cut down dramatically on out-of-stocks and reduce supply chain costs.

Among the technology areas P&G is most focusing on are data synchronization, electronic product code and radio frequency identification, advanced forecasting algorithms and even in-store devices that automatically track individual product movement and alert managers about potential out-of-stock situations.

Stephen David spelled out key areas of expected breakthroughs in the near future in an interview with WWDExecTech just before stepping down early last month from his position as chief information and business-to-business officer at the Cincinnati-based manufacturer. David retired from his full-time position at P&G but remains in a consulting role. 

One of the prime areas he cited as crucial for improving key areas of the business, including accurate delivery of product, is synchronized data. The issue has gained steam over the last two years as interest in RFID has surged. But most industry observers continue to question when, if ever, retailers and manufacturers will embrace data synchronization, a process that can be immensely costly, time-consuming and difficult to achieve and whose benefits are not necessarily immediately evident.

David is not one of those executives and, in fact, he sees adoption surging this year. “Today trading partners that represent 15 percent of our business are involved in data synchronization. We have pilots with Kroger in the U.S. and Metro, Carrefour and Delhaize in Europe. We are expanding broadly across all our categories. I expect that 15 percent to grow to 40 to 45 percent, which is our target for the end of 2005. That will certainly be a tipping point in making it a mainstream business process for P&G,” David said.

Asked why trading partners should get involved in data synchronization efforts, David cited estimates that $250 million of product in the United States is refused by retailers and returned to the manufacturer. “A big chunk of that, two-thirds, is because of bad data in the system,” he said.

This story first appeared in the February 15, 2005 issue of WWD. Subscribe Today.

No customer refuses product because the price is too high or too low. If the price is too high, they deduct it; if the price is too low, they accept it. Rather, retailers refuse product because it is the wrong product, and that probably also means an out-of-stock situation and lost sales, David said.

In examining the master data file of some of its key trading partners, P&G discovered that, under the best circumstances, 25 percent of all item files contained inaccurate information, and in the worst cases, as much as 70 percent.

P&G, for its part, has published nearly 30,000 sku’s to the Transora [the manufacturer oriented] global registry. “We have within our grasp and vision some big steps that trading partners can take to lower costs, to speed goods to market, to reduce counter-fitting, and to make transformational changes in supply chain and business processes,” David said.

Another aspect of data synchronization that is being overlooked, to some extent, is the fundamental role it plays in allowing other transformational technologies, such as RFID, to flourish.

Asked about frustration with regard to the pace of change, David replied, “My frustration is we have to do the basics such as developing business cases. Adequate-to-good work is under way, but some greater proportion of time has to be spent on the transformational potential an ePc environment could offer — getting paper out of the system, dramatically changing the way products are shipped and received in the system so people don’t have to count them, reducing the amount of deductions,” David said.

“We tend to have a myopic view even in getting market share information. It is done with 50 sets of mirrors, as opposed to starting today with the point-of-sale and setting up systems that will then naturally feed into an EPC environment within three to five years,” he added.

David said the industry must take the steps necessary to combat perhaps the biggest problem of all-out-of-stocks. Doing so, he noted, is both a technological and business process challenge. Studies continue to show that, of the top 2,000 items in grocery, for example, more than 10 percent are out of stock at any given time, and the same holds true for other retailing segments. 

A third of the problem resides in the store, according to P&G research; the product is in the store but not on the shelf. Another third of the problems are in the distribution center; product may be in transit or in the wrong rack but it is not where it is supposed to be. And a third can be traced to problems at the manufacturer, such as shipping incorrect orders.

“We have called this out as a corporate strategy. We call it the consumer-driven supply network, or CDSN. We have been working on it for about a year. We believe we need entirely new measures to not only measure out-of-stocks but also identify the causes of them on an individual, specific basis,” David said.

“We are also looking at the quality of our product on the shelf as never before. We know the quality of the product when it leaves the plant but what is the quality when it gets to the shelf? Does it have dust on it or is it dented? Sometimes we find 10, 15 or even 20 percent more degradation of product then when it left our distribution center. We need to find better ways to get it to the store faster so it doesn’t sit in the distribution system,” he added.

One device P&G has deployed in some tests, which was once dubbed heartbeat, “sits at the POS and looks for patterns. After a month or so of studying an individual store, it then might say, ‘Gee, I haven’t seen this particular sku of Tide in 20 minutes.’ That is out of pattern. It is probably an out-of-stock, and it alerts the store manager by sending a text message to a computer. We have seen dramatic reductions in out-of-stocks in the stores we have tested this in,” David said. 

Asked if the reduction in out-of-stocks exceeded 50 percent, David replied that figure “was probably a little aggressive but the results were very material. We have piloted it for more than a year, and one chain in the U.S. — whom I can’t name — is now rolling it out to a big chunk of their stores because they got such good results. When we can tie that together in an EPC environment, it is going to be even more effective,” he stated.

P&G is investing heavily — in time, money and resources — in the area of advanced forecasting. As one time, not so many years ago, the job of forecasting was typically allocated to a person at the end of his or her career and was held in fairly low regard as a worthwhile tool or benchmark to gain any business benefit. That has changed dramatically.

“We have done a tremendous amount of work in trying to get better at forecasting. This is an area that will continue to get better and better and better,” David said.

One thing the company is now doing is going to experts in supply chain at universities and have them help develop better forecasting algorithms because, David said, “at the sku level the variation per month could be 100 percent or even 200 percent versus what the forecast was. The internals didn’t look anything like what was being forecast.”

 That variability has been reduced dramatically by putting dedicated people into the role and having those people report to the unit business leader. “So you put in accountabilities, and you add systems and support, and now our forecasting is getting better,” David said. 

In part as a result of those changes, P&G has improved its line fills substantially. It has also beefed up its ability to manufacture products at a much swifter rate.

“Roughly speaking, if you dialed back 10 years at Procter & Gamble, we had the capability to produce any sku once per month. About four to five years ago, we started to be able to produce any sku once per week. What we are doing now and we are still in the process — we are better off in some products than others

— is getting to the point that, based on demand, we can produce any given sku any time of day.

“Now, we don’t do that. It is only based on what the need is. But that is our vision. Produce to demand,” he said.