March 20, 2008
Supply Chain Digest: The On-Going Battle over CPFR
By SCDigest Editorial Staff
Through a special arrangement, what follows is an excerpt of a current article from Supply Chain Digest, presented here for discussion.
It’s hard to think of any area of supply chain management that causes more
debate than Collaborative Planning, Forecasting, and Replenishment (CPFR).
Following a highly successful 1996 pilot aimed at keeping Listerine stocks
on the shelves of Wal-Mart, the VICS committee of the Uniform Code Council
(now called GS1) in the late nineties developed the famous “nine-step” model
that was released with much fanfare, and strong early support from retailers,
manufacturers, and supply chain software vendors.
What’s happened from there is a matter of some debate.
In the inaugural issue of CSCMP’s Supply Chain Quarterly magazine in mid-2007, consultant Rich Sherman wrote an article titled Why has CPFR Failed to Scale? In Mr. Sherman’s view, the basic problem is a disconnect between the objectives of retailers and manufacturers.
“What’s lacking is a reasonably straightforward capability to systematize the process of balancing and synchronizing retail replenishment requirements with manufacturers’ shipping and production requirements,” Mr. Sherman wrote. “As a result, CPFR implementations have been relegated to the strategic few — that is, the largest customer/supplier relationships — rather than scaling to encompass the entire market.”
In response, Joe Andraski, president and CEO of VICS, rebutted Mr. Sherman’s entire premise, “To the best of my knowledge, there has never been a negative CPFR article written by a member of VICS or anyone who has had direct experience with implementing CPFR.”
However, Jeff Harrop of Demand Clarity, said, “The scalability of the current CPFR business process has been ‘the elephant in the room’ for quite a few years now. The simple fact is that the highest levels of both out-of-stocks and finished-goods inventory in the CPG (consumer packaged goods) supply chain are at the retail store. In and of themselves, higher-level management processes like CPFR were not designed to deal with what goes on day by day, item by item, and store by store.”
Ed Nieuwenhuis, a former supply chain manager at retailer Meijer stores, also agreed with Sherman, but still sees great potential for CPRF.
“Our suppliers would refer to ‘forward buys’ and ‘diversion’; we would look at making ‘investment buys’ and collaborating with each other to get the biggest economic punch from the pricing programs offered by our suppliers,” Mr. Nieuwenhuis responded. “Only when the focus is on reducing total logistics cost can collaboration succeed.”
Robert Nardone, former senior supply chain executive at Unilever North America, told SCDigest that for consumer goods manufacturers with advanced Sales and Operations Planning (S&0P) processes, collaborative forecasting data from retailers or other customers is key to the planning and decision-making process.
“It’s essential to get good forecast information from retail customers to develop demand and supply plans,” he added. “That is happening with leading manufacturers, whether you call it CPFR or not.”
Discussion Questions: Is there a fundamental problem with the collaborative planning, forecasting, and replenishment (CPFR) model? Has it come close to meeting its potential? What are the real barriers to CPFR success?
BrainTrust
Recent Discussions
Discussion Questions
Poll
There is a disconnect between retail replenishment and the manufacturers’ shipping and production requirements, in part because accurate forecasting needs constant adjustment to match up with the changing realities of the in-store environment, external factors outside of the product and brand’s control that impact sales velocity on any given day, week, or month, and activities within the product category that have impact on sales.
CPFR involves many extremely complex business processes at a variety of levels in more than one organization. Evaluating whether CPFR “works” doesn’t seem to be the right question. Effectively implementing CPFR activity is a journey taking place in fits and starts by focusing on specific projects. Each time one project is completed there is another part of the process to address. Keeping everyone trained and focused is an huge challenge.
CPFR is not a quick fix for anything or a one time project. If approached that way it is better not to begin. To be effective, CPFR needs to be approach as a long term process which begins with a single activity or focus involving constant training and communication at every step.
I’m struck by what Frank Dell wrote about different agendas. Both retailers and manufacturers have an interest in inventory pooling up, only not on their books.
For years, retailers used their relative leverage to assure inventory-on-demand that’s held by their vendors. With the increasing consolidation in retail, this really puts manufacturers behind the eight ball. Conversely, before the consolidation really got underway two decades ago, manufacturers used their relative leverage very effectively to push inventory down the supply chain onto retailers.
Each is seeking to maximize sales without assuming the inventory risk and the carrying costs. Inventory inevitably pools with the weakest link in the supply chain, whether it be manufacturer, importer/marketer or retailer. This is a powerful disincentive to cooperation.
I did an in depth interview of the stakeholders (supply chain folks, category managers, mass+food, CPG) for a VC firm interested in this space weeks after the Listerine demo at the Kennedy School that was run by Benchmarking Partners–I recall at that time the overwhelming concern was the level of work required by the retailer (by retailers) to make CPFR work–I also recall that the initiative had great similarities to the vibe of the RFID mandate kickoff meeting at Bentonville, namely: a sophisticated few with staff and leadership to make it happen vs. the rest. That was ’96, so things may have changed but wanted to throw that out for what it’s worth.
There are three problems with CPFR. It does not work, as the two parties have different agendas, needs and issues. The second reason it is an obsolete approach due to technology advances. The third reason is the use of a forecast. There is simply no reason to waste time on forecasting store and chain movement today.
The Demand Management approach uses POS item movement and with this the complete supply chain from store replenishment through raw material procurement can be driven and managed. Knowing what was sold at retail today tells one what to schedule for production a few days out.
What Frank Dell said!!!
CFSP works, but not as an isolated life cycle. It is dependent on timely, outside aggregate information to define and develop its efficiencies. Despite all of this it is still a simple snapshot in time, and does not represent a dynamic, moving, supply chain solution that is continually adjusting for the nuances in the retail environment at the store level.
CFSP is only as good as the information it has, and needs to be treated as a solution to represent this information, when it was taken. The robustness of the information will be directly reflected in the proposed solution, and the ability to implement this will be reflected in the capabilities of the supplier to properly meet these within the given framework and timeframe of the CFSP.
Here’s what will break the back of retailer data protection: consumer demand. Consumers are increasingly a part of this chain through the internet. Site-to-store purchasing and consumer inventory checks (such as NearBy Now and JC Penney) will create traffic where there is supply. This “near perfect” information means that consumer demand for available inventory will actually influence retailers in real time. That’s a game changer. When replenishment becomes a competitive advantage (and I believe that day is nearby now), watch retailers open up.
So much progress could be made if both parties were truly invested in the right outcome, but far too rarely does that happen. As has been pointed out, too many retailers want to sell or protect their data from their suppliers. That also “protects” them from the benefits. Manufacturers want to supply the data and resources that will only help their brands and not the category. Who can blame them on one hand but it doesn’t create trust on the part of the retailer. As long as protection is more important than savings, things will not get any better any time soon.
I have argued for years that CPFR should actually be “CPFVMI.” The process works well enough for planning and forecasting, because it’s designed to handle the negotiative part of “my objectives vs. your objectives”–and actually several software companies have tackled streamlining this process with automation. Where it breaks down is with the “R.”
Replenishment needs to be high-visibility, high-speed. It is the safety valve that lets both parties react to unexpected demand and the natural forecast error that will never go away. Retailers need to set stocking objectives–service levels–give manufacturers real-time item movement data at the store-level (oh wait–many want to charge for that! Some only provide it aggregated! Some provide it too late to do anything about it!)–and then let the manufacturers figure out the lowest cost way to meet their service levels.
Ironically enough, if the industry did that, everybody stands to sell more stuff, and thus make more money.
Interesting article. There are two factors to having success with this valuable program. It’s collaboration and to have collaboration you must have trust. The second is technology.
Many companies are hesitant to share what they perceive as proprietary data. The other issue is that technology affords the ability to have and use good quantitative data yet many internal software systems have not been upgraded or designed to take advantage of the information analysis, visibility or process improvement for proactive results.
CPFR should be easy for those retailers who are also their own suppliers. If a store is 100% private label, like Apple or Old Navy or H&M or IKEA, everyone’s allegedly on the same team, aligned by the common source of their paychecks. But everyone familiar with brand-controlled retailers knows “it just ain’t so.” Folks have trouble working as an aligned team with people 2 desks away in the same company, let alone miles away, in multiple companies.
More Discussions
Has Same-Day Delivery Become an Expectation Across Retail?
Are expectations around retail delivery speed in flux?OpenAI Names the Infrastructure Standard for Agentic Commerce
Will OpenAI or Google (or another entity) set the gold standard for agentic commerce?Is Five Below Poised For Even Greater Growth?
Five Below is hitting serious growth numbers. Can this trend continue?
There is a disconnect between retail replenishment and the manufacturers’ shipping and production requirements, in part because accurate forecasting needs constant adjustment to match up with the changing realities of the in-store environment, external factors outside of the product and brand’s control that impact sales velocity on any given day, week, or month, and activities within the product category that have impact on sales.
CPFR involves many extremely complex business processes at a variety of levels in more than one organization. Evaluating whether CPFR “works” doesn’t seem to be the right question. Effectively implementing CPFR activity is a journey taking place in fits and starts by focusing on specific projects. Each time one project is completed there is another part of the process to address. Keeping everyone trained and focused is an huge challenge.
CPFR is not a quick fix for anything or a one time project. If approached that way it is better not to begin. To be effective, CPFR needs to be approach as a long term process which begins with a single activity or focus involving constant training and communication at every step.
I’m struck by what Frank Dell wrote about different agendas. Both retailers and manufacturers have an interest in inventory pooling up, only not on their books.
For years, retailers used their relative leverage to assure inventory-on-demand that’s held by their vendors. With the increasing consolidation in retail, this really puts manufacturers behind the eight ball. Conversely, before the consolidation really got underway two decades ago, manufacturers used their relative leverage very effectively to push inventory down the supply chain onto retailers.
Each is seeking to maximize sales without assuming the inventory risk and the carrying costs. Inventory inevitably pools with the weakest link in the supply chain, whether it be manufacturer, importer/marketer or retailer. This is a powerful disincentive to cooperation.
I did an in depth interview of the stakeholders (supply chain folks, category managers, mass+food, CPG) for a VC firm interested in this space weeks after the Listerine demo at the Kennedy School that was run by Benchmarking Partners–I recall at that time the overwhelming concern was the level of work required by the retailer (by retailers) to make CPFR work–I also recall that the initiative had great similarities to the vibe of the RFID mandate kickoff meeting at Bentonville, namely: a sophisticated few with staff and leadership to make it happen vs. the rest. That was ’96, so things may have changed but wanted to throw that out for what it’s worth.
There are three problems with CPFR. It does not work, as the two parties have different agendas, needs and issues. The second reason it is an obsolete approach due to technology advances. The third reason is the use of a forecast. There is simply no reason to waste time on forecasting store and chain movement today.
The Demand Management approach uses POS item movement and with this the complete supply chain from store replenishment through raw material procurement can be driven and managed. Knowing what was sold at retail today tells one what to schedule for production a few days out.
What Frank Dell said!!!
CFSP works, but not as an isolated life cycle. It is dependent on timely, outside aggregate information to define and develop its efficiencies. Despite all of this it is still a simple snapshot in time, and does not represent a dynamic, moving, supply chain solution that is continually adjusting for the nuances in the retail environment at the store level.
CFSP is only as good as the information it has, and needs to be treated as a solution to represent this information, when it was taken. The robustness of the information will be directly reflected in the proposed solution, and the ability to implement this will be reflected in the capabilities of the supplier to properly meet these within the given framework and timeframe of the CFSP.
Here’s what will break the back of retailer data protection: consumer demand. Consumers are increasingly a part of this chain through the internet. Site-to-store purchasing and consumer inventory checks (such as NearBy Now and JC Penney) will create traffic where there is supply. This “near perfect” information means that consumer demand for available inventory will actually influence retailers in real time. That’s a game changer. When replenishment becomes a competitive advantage (and I believe that day is nearby now), watch retailers open up.
So much progress could be made if both parties were truly invested in the right outcome, but far too rarely does that happen. As has been pointed out, too many retailers want to sell or protect their data from their suppliers. That also “protects” them from the benefits. Manufacturers want to supply the data and resources that will only help their brands and not the category. Who can blame them on one hand but it doesn’t create trust on the part of the retailer. As long as protection is more important than savings, things will not get any better any time soon.
I have argued for years that CPFR should actually be “CPFVMI.” The process works well enough for planning and forecasting, because it’s designed to handle the negotiative part of “my objectives vs. your objectives”–and actually several software companies have tackled streamlining this process with automation. Where it breaks down is with the “R.”
Replenishment needs to be high-visibility, high-speed. It is the safety valve that lets both parties react to unexpected demand and the natural forecast error that will never go away. Retailers need to set stocking objectives–service levels–give manufacturers real-time item movement data at the store-level (oh wait–many want to charge for that! Some only provide it aggregated! Some provide it too late to do anything about it!)–and then let the manufacturers figure out the lowest cost way to meet their service levels.
Ironically enough, if the industry did that, everybody stands to sell more stuff, and thus make more money.
Interesting article. There are two factors to having success with this valuable program. It’s collaboration and to have collaboration you must have trust. The second is technology.
Many companies are hesitant to share what they perceive as proprietary data. The other issue is that technology affords the ability to have and use good quantitative data yet many internal software systems have not been upgraded or designed to take advantage of the information analysis, visibility or process improvement for proactive results.
CPFR should be easy for those retailers who are also their own suppliers. If a store is 100% private label, like Apple or Old Navy or H&M or IKEA, everyone’s allegedly on the same team, aligned by the common source of their paychecks. But everyone familiar with brand-controlled retailers knows “it just ain’t so.” Folks have trouble working as an aligned team with people 2 desks away in the same company, let alone miles away, in multiple companies.