Is Sales Forecasting an Art or a Science?
Posted on February 7, 2017By John Kimball, Senior Consultant, Lanham Associates
It’s easy to assume the sales forecasting process is very people-dependent, and that it’s very much an art rather than a science. But what if the reverse were true? What if you could put management rules in place to support management thinking and strategy? Here are five management rules you may want to consider adding to your forecasting process, plus one great thing you can do for your business in order to get even more from your forecast.
#1. Use the formula that works best for each item. Using computer simulation, rather than manual calculations, to go back in time and compare what would have been the forecast error for different formulas, had each been chosen, is a very good example of how to use technology effectively. Provided the sales history is ready for the forecast process (see Getting Sales History Right for Forecasting), few can argue that the selected formula and its forecast are not reasonable.
#2. If sales exceed the forecast for a month, revise the forecast immediately. Using simple management strategies like those below, you can define ahead of time how to handle this:
- If sales exceed forecast for the month, increase the forecast to equal total sales.
- If sales exceed forecast for the month, increase the forecast to equal total sales and add x percent to the forecasted daily rate for the remaining days of the month.
- If sales are exceeding the daily forecast rate, add the increase to each day’s forecast for the number of days that have passed in the current month.
#3. Add specific amounts to the forecast for special events/promotions. Rather than simply increasing the forecast for a one-time event or promotion, you can add quantities to the forecast of certain items for the exact time period it is planned, along with your notation with the reason for the expected increase. This allows you to manage the increase before the sales occur, and the histories after the sales occur.
#4. Work collaboratively to create a forecast with major customers for major items. Identifying dominant items with select customers, you can work with them on each item to share the sales histories, your best fit formula/forecast, and mutually agree on a forecast without disrupting the forecast for the balance of your items. Now you can focus on why the forecast is what it is and you’re relieved from having to calculate the forecast itself.
#5. For peak period seasonal items use a rule to increase safety stock during this period. Rather than reviewing all seasonal items and calculating a safety stock for each, it is much simpler and more effective to identify the seasonal items, the periods the season occurs, and set an increase in the number of periods of safety stock your management strategy supports. This allows you to focus on managing to a defined strategy without being encumbered with the clerical work.
Shorten lead times from suppliers. With all the time that has been freed up by doing 1 through 5 above, you will have time to work on reducing lead times from suppliers. Once lead times are reduced, having a forecast error is less of a problem because your suppliers can respond much more quickly to your customer needs. Read 10 Lead Time Reduction Tips for more on how you can do this!
If you’re interested in learning more, check out our Advanced Forecasting and Procurement solution.
