Dr. Jain, editor of the Journal of Business Forecasting, answers readers’ questions on forecasting and demand planning.
[Q] I’m working in the retail industry (home center) and it’s a challenge to track the performance of new products. Would you recommend any process for that?
[ A ] The best thing is to look for one or more similar products that were launched in the past. Determine from them what percentage of their total sales come in the first, second, third, and fourth week, and use those percentages to determine the sales you are likely to get from the products you recently launched. This will give you some indication how a given product is performing. If you don’t have data on products that are similar to yours, then use ones that are close enough.
According to Rick Davis, Vice President and Global Lead, Office of Data Acquisition and Governance at Kellogg, 70% of the first quarter sales come in the first three weeks. This can also be used as a guide about how a product is likely to perform. Regarding the process, I think it’s important to have a dedicated person or team that regularly monitors the performance of new products, makes a proposal, and then presents it to the leadership. Once approved, implement it right away. Whatever action is taken, it has to be communicated to the S&OP team so that it can incorporate it in its overall plan. How often the team should meet depends on the life cycle of a product, as well as the risk involved. For products with a shorter life cycle or for ones that are high in value and risk, it may be better to meet weekly.
[ Q ] I am trying to understand if our growth rate is increasing or decreasing. I have a series of daily data for two years, and am struggling to find ways to calculate the growth rate at a monthly level and express it in a way the management can easily understand. Any ideas?
[ A ] Here are the steps you can to take to calculate the growth rate. Step 1: Convert the daily data into months. Step 2: Compute the rate of growth from one month to the next, that is, the rate at which sales increased from one month to the next. Step 3: Develop a graph using Excel from the growth rates, and then right click on it. One of the options that will pop up is “add a trendline.” There are different types of trend lines; pick the linear one. If the line is moving upward, the growth rate is increasing, and, if it is moving downward, it is decreasing.
[ Q ] What is the difference between scenario planning and risk management?
[ A ] The scenario planning, also called “what if analysis,” is a tool used to find a solution that maximizes revenue and profit and minimizes risk. In the S&OP process, it is used to manage demand and supply. The demand planners may ask various questions to make a best-policy decision. What would happen if we lower or raise the price by a certain amount, lower or raise the advertising budget by a certain percentage, launch one or more products, and so on.
To manage supply, supply planners may go through various scenarios to determine how much inventory they should hold, which would minimize shortages and surpluses as well as risk, and optimize profit and service level. To manage customers, planners may ask questions such as this: what would happen to the total revenue and profit if we get rid of some low-performing customers? In each case, the objective is to go through different scenarios to find the best solution—a solution that optimizes revenue and profit and minimizes risk.
[ Q ] Do you have any information regarding how to dollarize forecast accuracy? I work for a $1B company and I’m wondering how many dollars will be realized with each percentage improvement in forecast accuracy by our Demand Planning team. I heard somewhere that the savings is $1M for every 1% for a company of my size.
[ A ] How much you can save depends on the size of a company, as well as on how developed the forecasting process is. In my study of Consumer Products Companies, I found an average company can save $3.52 mil. for every one-percent improvement in the under-forecasting error, and $1.43 mil. in improving the over-forecasting error. This saving is only from the improvement in the supply chain. It will be much more if we account for savings resulting from other uses such as in sales, marketing, and strategic planning, all of which are hard to measure. There is a tool on the IBF website where one can calculate savings resulting from forecast improvement by putting in one’s own numbers.
These questions and answers were published in the Spring 2017 issue of the Journal of Business Forecasting, the industry-leading academic journal for the forecasting and demand planning disciplines. The questions were submitted by readers and answered by the editor, Dr. Chaman Jain. Become an IBF member and receive your subscription to the Journal of Business Forecasting.