Here is the final part my interview with P.J. Jakovljevic from Technology Evaluations Centers (TEC) on sales and operations planning (S&OP). If you missed them, check out part one, part two, and part three. The entire Q&A along with PJ’s introduction and commentary on Kinaxis can be found here (free registration required). PJ: Do you see a link between S&OP and multi-echelon inventory optimization (MEIO), and what do you offer in that regard? TM: There is a wider link between S&OP and multi-tier visibility, planning, and control. S&OP has been traditionally seen as an internal process focused on getting consensus among different internal functions. But S&OP was conceived well before the advent of pervasive outsourcing and offshoring: a time when the manufacturing, let alone commodity management and procurement, were just ‘down the hall’—a time when most customers were in the same country and spoke the same language. The intervening years haven’t reduced the level of visibility needed. In fact, offshoring has increased the need for multi-echelon visibility, while outsourcing has reduced multi-echelon visibility. So I am absolutely convinced that multi-echelon inventory visibility, and of course demand visibility, is closely linked to S&OP. How inventory levels will be set will depend on the contractual agreements between the original equipment manufacturer (OEM)/brand owner and the contract manufacturer. In addition, S&OP is often referred to as sales inventory operations planning (SIOP) and production, sales, inventory (PSI)—both cases emphasizing the inclusion of inventory management in the overall S&OP process. RapidResponse provides ways to determine ideal inventory levels given the customer service level targets and historical demand and supply patterns. Scenarios can then be used to test the financial and operational consequences of changing safety stock or reorder point (ROP) values. PJ: There is indeed a great deal of cross-functional cooperation and collaboration that is required for managing S&OP. How are companies enabling this, and are they doing it successfully? TM: From a process perspective, it is difficult to get people to work across functional boundaries, let alone across organizational boundaries. In addition, technology is often a barrier when each function has its own data and systems for analysis. Not only are there arguments about the results, but also about the data. Another barrier is that often, particularly in outsourced environments, people don’t even know whom to call in another function or company to resolve an issue. As discussed above, we enable cross-functional cooperation and collaboration by providing a solution that has a single data model, a single set of analytics, and a single UI. Naturally, we have a full security model and the ability to filter and aggregate the data to make it relevant to the person and their role. If anyone makes a data change that has a significant impact on someone else, the person affected will be alerted immediately and can then collaborate on resolving the issue. Naturally, more than two people can collaborate in a given scenario. Any changes are immediately visible to all participants in the scenario. PJ: What is your take on the link between S&OP and collaborative planning, forecasting and replenishment (CPFR), and what are your customers doing in that regard? TM: CPFR, in the strictest terms, has been a failure largely because it was a burdensome process and because of the expectation on the part of consumer packaged goods (CPG) manufacturing companies that it would improve the forecast accuracy of the retailers. If instead we take CPFR to simply mean a more collaborative and inclusive planning process between trading partners, then I would say it is on the rise. As already discussed, outsourcing of manufacturing has led to the need for CPFR between the OEM and the contract manufacturer by extending internal cross-functional cooperation and collaboration to contract manufacturers in particular. PJ: If you had to name the top three priorities for a company looking to evolve their S&OP process, what would they be? TM: I would say the following:
- Do S&OP more frequently, preferably continuously.
- Do S&OP collaboratively and consecutively, not sequentially through a traditional five-step process.
- Understand that S&OP is one step in a planning continuum, and that all steps need to be synchronized constantly.
All planning, let alone S&OP, should be governed by the principle that execution against the plan should be monitored continuously. In addition, if there are major market shifts, a company must react quickly by regenerating the S&OP plan. But exception management is important not only in monitoring execution of the plan and market shifts, but also in detecting big changes in the plans being generated. But the exceptions need to be relevant to the person receiving the alert, and the supporting data needs to be packaged in a manner that is relevant to that person. PJ: How and where do ‘what-if’ capabilities fit into the S&OP process? Is it a priority capability for an effective S&OP process? TM: I don’t see how S&OP can be carried out effectively without strong ‘what-if’ capabilities. It is more than a priority—it is a core requirement! Humans are so much more creative than machines. What they need is a rapid way of testing alternatives and evaluating the consequences in a timely manner. If they have an effective manner to understand the effect of the decisions on financial and operational metrics, they will inevitably make the right decision, particularly when several alternatives can be compared side-by-side. PJ: What is the role of master data management (MDM) in S&OP, and what is Kinaxis doing in that regard? TM: Because of our long history in the outsource high-tech/electronics space, we have had to deal with MDM-like issues for a long time, particularly equivalent item numbering, including bill-of-materials (BOM) structures. We focus a lot of attention on data quality and have a number of workbooks that identify missing and incomplete data. Without a doubt, MDM systems do increase the quality of data, but I do not see any reason to delay the deployment of RapidResponse until the MDM system is in place. Clearly, the business is being run with the existing data, and using the existing data in a more effective manner would only be beneficial. However, I would not recommend waiting for perfect data. If an MDM system or data warehouse already exists, we can integrate it. PJ: Some S&OP/IBP players offer functionality (often via acquisitions) for DP, trade promotions, financial consolidation, strategic network optimization (SNO) (‘what-if’ simulations of networks), and even PLM capabilities for NPI/PPM. What is your plan of action for successfully competing with these much broader and strategic-level S&OP offerings? TM: As stated above, we provide a single solution to satisfy many supply chain processes using a single data model, a single set of analytics, and a single UI. We believe we already have a broad and strategic-level offering. Our analytics either currently covers all of the capabilities you list above or will shortly. We have a fundamental issue with the concept that S&OP can be satisfied effectively using an overlay solution that has a separate data model and analytics from other planning tools. How can NPI be separated from DP/forecasting, supply planning, and capacity planning? By extension, how can you finance these other functions? The relative importance of these adjacent capabilities will depend on the industry, and clearly we have broader coverage and deeper capabilities for the industries we focus on, namely high-tech/electronics, aerospace, industrial, and pharmaceutical. PJ: Do you have any other observations and trends related to S&OP that haven't been mentioned in the previous questions? TM: Without a doubt, there is a trend to a broader description of S&OP, particularly in outsourced environments where the contract manufacturers, at the very least, need to be included in the S&OP process. But, this is not only broader in ‘geographical’ coverage, but also in departmental function and time horizon. That's it! Thanks for following along.