New Links for a Winning Supply Chain

It was Peter Drucker, the original management thinker, who once wrote: "Manufacturers normally want to make what they want to sell, rather than what consumers want to buy." Today, a relatively new set of theorists is endeavouring to resolve that paradox through a process they call "collaboration"--the sharing and agreeing of plans and forecasts over the Internet to try to predict and fulfil customer demand.

Their theories centre on the concept of the demand chain: the idea that demand from consumers can be passed back through the supply chain to a retailer's immediate suppliers, and on to their suppliers, until the whole is working in real time to satisfy customer needs. However, in spite of an enthusiastic take-up by early adopters who reported increased sales, and an obvious increase in customer satisfaction, some confusion remains over what is real collaboration and what is merely old-fashioned supply chain management.

The idea of passing on details of consumer demand may seem no more than common sense, but according to Joe Andraski, chairman of the Collaborative Planning, Forecasting and Replenishment (CPFR) project (www.cfpr.org), it is the final piece in a jigsaw that has been evolving in manufacturing for the past three decades. "It is a continuous evolution. Everything that has happened in supply chain management has been an improvement. Companies focused first on transportation, then looked at their warehousing, and that came together in distribution. Then they said: 'let's look at our purchasing, our suppliers' purchasing and our suppliers' suppliers.' Then they realised if it wasn't all synchronised, it wouldn't work. As we've continued to evolve it, we've gradually brought it all together. It has been the catalyst for what we are doing today."

Andraski is well placed to discuss the evolution of supply chain management. For 24 years, he assembled a supply chain organisation at snack maker Nabisco that was consistently rated number two behind Procter & Gamble in the annual survey of Grocery Manufacturers' Association members. Given Nabisco's size, $2.5bn compared to P&G's $30bn, this was some achievement. Four years ago, he became involved in CPFR, framing a set of guidelines for collaboration between trading partners.

Andraski believes CPFR marks a significant step forward in supply chain management. He believes the idea of collaboration itself is coming of age, thanks in part to the Internet. "Everyone now understands that collaboration is not a necessary evil. It isn't consorting with the enemy." As well as the numerous digital exchanges springing up every day, he cites the example of U.S. supermarket chain BestBuy, which shares its strategic plans with manufacturers on a store-by-store basis. "When you ask them whether that is sensitive information and what risks are involved, they reply that the risk is worth the reward."

At Nabisco, Andraski says the key enabler was the buy-in from management, which gave the supply chain initiatives their full support while the company embarked on a massive project, building bespoke functionality around the core order management and inventory management systems of the Seventies. Only recently has the company implemented planning and scheduling capabilities from packaged supply chain vendors, including Manugistics and i2.

Bottom-up Ideas

It was a fantastic proving ground for ambitious young supply chain practitioners such as Andraski, and he believes that assembling a team which pulled ideas from the bottom up was second only to the management drive. "We built organisational capabilities around the technology and then bred a culture. It was externally focused: we would share a significant amount with the outside world--in the grocery industry and beyond - because that attitude allowed us to keep up with developments. We were very in tune with customer requirements both inside and outside the industry."

Nabisco's IT people were also in tune with the business. "We were fortunate in having a systems organisation that built what we wanted and not what they wanted. We didn't have to fight through every specification and modify our needs for the system. At meetings, you couldn't figure out who was the systems guy and who was from the business."

Nonetheless, after being involved in the efficient consumer response (ECR) initiative for a couple of years, Andraski began to feel that there was something missing. He had led the writing of best practices for continuous replenishment (CR), the idea that once an item was out of stock it would already have been replaced, and Nabisco was an early adopter, but after seeing the work CPFR was doing, Andraski realised CR was not enough.

"CR was a great stepping stone, but it was not achieving the overall objectives for a number of reasons. In terms of the resource, it was very manufacturing-centric, yet the benefits all fell to the retailer. Plus it could never fulfil the promise of getting to a critical mass [of users], because there was no forecast involved."

When a CR analyst created a plan, mapping out customer orders, Andraski felt its strength still depended largely on their skill: "Their knowledge, their ability to work with the organisation, the strength of the marketing. Often the replenishment order was more to do with the past ten years of doing business than a forecast." The problem was that in the fluid consumer market, the following year's business may bear little relation to the past ten. Competition, seasonality, fashions, promotions and price changes all have dramatic effects on the shape of future business, but analysts were operating in a vacuum, insulated from the people who knew most about it - the trading partners. 

Wild Success?

There was clearly a need to bring partners together to formulate an up-front plan, turn that into a sales forecast and then come up with an order forecast. CPFR began a series of prototype test cases to prove this could work. Andraski claims they have been wildly successful, and a host of global companies have signed up to the initiative, with five pilots identified (see box next page). Although few are talking beyond pilots, he says that a number have claimed in private that the overriding benefit is increased sales.

"It is another business opportunity. We have heard from supply chain management of increasing sales efficiency and effectiveness, but CPFR also brings increased sales. If you can manage your business with less inventory and resources and sell more, [then you can influence] the measures that CEOs live by."

The real competitive advantage, according to Andraski, comes not from a technology medium but from the ability to plan in advance. He claims most companies spend so much time looking at short-term problems that the potential to agree a plan and then manage by exception, only revisiting it when a major irregularity is flagged, is a breath of fresh air. "Between 30 and 80 per cent of companies' resources are spent fighting fires--looking for a lost order, correcting a billing or trying to meet a forecast they're not signed up to. They claim 30 per cent, but secretly admit it's as much as 80 per cent."

However, all is not as rosy as it might seem in the CPFR garden. Wegmans, a Nabisco customer, was a case in point. The supermarket chain embarked on an early CR pilot with Nabisco for its Planters peanuts range, but found its technology capability insufficient to generate an outline forecast. It has since invested further and joined CPFR, and now reports sales up significantly.

IT investments aside, CPFR may be pushing its usefulness too far in its efforts to cover all bases. "The principles are all the same," insists Andraski, and it has won endorsements from different sectors of manufacturing. The hi-tech RosettaNet consortium has adopted CPFR guidelines in its recently published supply chain standard for the IT industry, and this is beginning to make its way into supply chain products. The clothing consortium DAMA (Demand Activated Manufacturing Architecture) was also a founder member in CPFR, playing a major role in defining the guidelines.
Andraski says an executive from automotive giant Ford, which has embarked on its own very public supply chain initiative recently, was at a CPFR conference and attended all six presentations. "There is so much for me to learn here and take back to the company," he told Andraski. "With three, four or five echelons of suppliers, anything you can do to avoid the murky waters of misinformation is a plus," he says.

Univeral Appeal

However, the fact that organisations like RosettaNet have had to tweak the guidelines illustrates that it is not universally applicable. "It has made slight modifications and taken it a step further," says Andraski. "You can tweak it. Obviously, a wholesaler does business with a retailer in a different way to a retailer with its customers."

RosettaNet has also published a version of its collaboration standards in eXtensible Markup Language (XML), the new language of the Internet, which opens a potential can of worms for CPFR. Its original prototype back in 1997, with Sara Lee and Wal-Mart, was defined for electronic data interchange (EDI), the standard for communication back in the Eighties.
But while colleagues on the standards councils have been condemning the old standards as dead in the water, Andraski is more conciliatory. "There's no magic to this, it's just common sense. The importance of defining a standard is that it has to be around the technology. If Wal-Mart is doing business, it does not want to deal with four different standards. It would much rather do business in one way, because that is a lot easier to manage."

Meanwhile, other bodies and software companies have been attempting to hijack the collaboration juggernaut. Software vendors have started talking about collaborative commerce, for example, and end users have set up collaboration projects in isolation. In the UK, J Sainsbury has its own project, part of its Information Direct initiative, and has cited savings of £5m, but while Andraski claims this for CPFR, the supermarket makes no mention of the U.S. project.

Andraski is sanguine. "If the principles are there, I don't care what they call it. There's a certain pride of authorship around this, and software companies are all looking for their unique toehold, but if they end up with the same results and manufacturers don't have to set up a number of systems, it's OK."