The IBF Demand Planning & Forecasting conferences are always enlightening, including the latest event in San Francisco. While the demand planning vendors touted their latest statistical forecast algorithms and growing S&OP functionality, and customers shared stories about their new forecasting process, I kept seeing a soft spot in the demand planning process. This weak link was preventing many companies from achieving “best-in-class” demand planning performance.
Despite the ever-sophisticated demand planning algorithms, reporting and best practices, it struck me that customers still had a tremendous pain getting sales to provide a good forecast and demand planning system were not adequate. Here’s how some of the attendees described how they obtained the sales forecast:
“Sales always has the latest forecast insight and so we give sales a voice in our demand planning process by having them email us key account updates every month.”
“We take CRM opportunities from Sales and get a rough short-term forecast by taking the quantity times probability, as a weighted forecast. This gives my demand plan a reality check in case there’s a trend difference.”
“We initially planned to have the Sales team input a forecast in our demand planning system but reverted back to spreadsheets. At any rate, sales folks can be very accurate in their forecasting.”
I have heard comments just like these many times in my career. Because of my passion for forecasting and planning, I have worked at leading-edge companies like i2, Manugistics (both now JDA), Steelwedge, and now Right90. My views have been informed by working with 50 or so companies across many industries. I have seen the demand planning processes that work with Sales effectively, but more often than not, they work not-so-well. Demand planning systems from Demantra, JDA and SAP APO DP are complex and powerful but not designed for the busy sales team.
Companies have also used statistical forecasting solutions for the past 35 years (e.g. Autobox), yet those have not solved the problem either. Effective sales forecasting solutions are only now emerging. This is an encouraging development, as IDF customers know the value of a sales forecast, and the value of including it in the demand planning process. After all, sales is closest to customer, and if their input wasn’t valuable, why would they have gone to all the trouble of manually including sales inputs from email and spreadsheets?
As a friend of mine says, “There’s a pony in there somewhere”. I’d like to explore how to get that pony out and on the racetrack. I’ll start by exploring the optimal process for connecting sales forecasting to demand planning and S&OP. As always, your thoughts are very welcome in this journey of discovery.
Tags: Best Practices, demand planning, sales forecasting, Statistical Forecasting