In nearly every conversation I have about planning, especially regarding the sales and operations planning (S&OP) demand consensus meeting, I speak to the need for critical thinking and active challenges. These are not throwaway expressions but key elements of a mindset that must be adopted in order to yield the best possible forecasting results. Critical thinking means using facts in decision-making. Active challenges describes a technique used to test, vet and dissect the assumptions — the subjective or non-fact-based criteria — used in planning. Both critical thinking and active challenges hinge on data, and a demand consensus meeting should leverage both extensively.
Consider for a moment the following inquiry in the context of a consensus meeting: “The forecast for blue widgets is +7 percent for the balance of the year, while the point-of-sale data and shipments are flat. Is there something we are missing in the data? Was there a change with our competition, or have we added some different promotional activity? The +7 percent change just does not make sense!”
As you can see, these questions serve a purpose. They are challenges to the demand plan based on data and directed toward uncovering an explanation. This is a prime example of actively challenging forecast data and assumptions via a process of critical thinking and examination.
At Combe, we make extensive use of data in the S&OP demand consensus meeting. Regardless of whether we are talking about a Vagisil crème offering or a Just for Men hair coloring shade, we bring to each meeting shipment trends, point of sales trends, events from prior years that may have influenced shipments, trade inventory changes, future events, and any other special activity such as IRC or displays. We look for harmonization across these indicators to help us verify our assumptions in developing a future demand plan. For example, if a Vagisil regular-strength crème is trending at +3 percent in year-over-year shipments, we look for similar trends in the point-of-sale data, and we expect the forward forecast to be more or less consistent with this trend unless there are promotional events that might shape future demand. Now, while this example is specific to health and beauty products, the need — the quest for meaningful demand signals — is not unique to consumer goods.
Years ago, I consulted for a company that made construction adhesives used in the manufacturing of plywood. We looked at point-of-sale data for plywood from retailers such as Lowe’s and Home Depot, but we also looked at housing starts as a bellwether of future demand, and we openly discussed hurricane season projections and activities. We aimed to create a mix of demand signal data that would help us predict the forecast for the stable business while trying to hedge inventory based on potential (and far more uncertain) outcomes such as hurricanes. To that end, we usually included built-in inventory for storms as an upside hedge — even more so when meteorologists predicted an active hurricane season. The inputs were different than those in my Combe example, but the critical thought placed on demand drivers during the consensus meeting was nearly the same.
When I consulted for the tire industry, we used shipment data, including backorders and point-of-sale data from major tire shops, in the consensus meeting. We also brought in other extrinsic data pulled from automotive manufacturers, including projections of their future demand for various tire sizes, trends in tread type, and tire profile and width. We then aligned this data with historical tire-wear-out rates from the aftermarket and used this framework as the basis for pegging demand. This deep dive into demand drivers was very important because the production capacity required for different tire types took a considerable amount of time and money to construct. Thus, leading indicators of any sort became very important to demand planning and S&OP processes. In this case, the demand drivers were completely different, and the unstated goal was more about aligning demand with future (and very expensive) capital requirements, but the effort and the use of data within the consensus process was very similar.
As these examples illustrate, the critical analysis needed to arrive at a meaningful demand plan can sometimes be exhaustive, as well as exhausting. The effort required can go as far as estimating the amount of trade inventory a retailer might carry or looking at historical order patterns by retailer to see if the data reveals any seasonal demand patterns. There is no limit to the types of data one can bring into a consensus meeting; however, the true challenge will always remain finding the best and purest demand signals upon which decisions about your forecast can be made.
Gain deeper insight in S&OP demand planning during Pat Bower’s live presentation at our annual Best of the Best S&OP Conference in Chicago, June 14-15. Save $50 when you register with code BESTSOP