Jan 14, 2020
How to produce an accurate inventory forecast.
If you are not sure of what your sales are going to be tomorrow, how can you determine what you need to buy today? By being able to forecast your demand accurately means being able to make better ordering decisions. Having the right tools in place to help you with your demand forecast results in optimal inventory planning.
Even if you don’t have a demand planning system in place, all companies forecast their demand to some degree and may use one of the following to do these forecasts.
- Using your last x months sales figures
- Calculating on your min/max levels manually
- Using gut instinct to make buying decisions
There are many tools available to assist you in creating the best possible forecast without having to review every single item manually.
How to get the most accurate demand forecast?
Experience has shown us that a robust, structured forecasting process is critical. The benefits of improved forecasts include a reduction of slow-moving and obsolete stock, as well as a reduction in stock-outs. By having a healthier, more balanced inventory means significant savings.
What does a forecast process look like?
There are monthly and weekly activities. Let’s start with what should be performed every month.
At the start of the month, the system forecast should be reviewed, and all known information should be included in the forecast. The steps to achieving this include:
Use a forecasting engine to do the grunt work and create computer-generated forecasts for all items by :
- Using sales or demand history to generate the forecast.
- Picking up on seasonality, trends, intermittent demand, one-off sales spikes, and factor in data such as lost sales.
Any forecast engine worth its salt should generate forecasts using several different algorithms. It should then compare all of those generated forecasts with the sales/demand history to determine the “best fit” forecast. This process covers the bulk of your items, leaving a small percentage of items that will need manual intervention. No forecast engine will get every forecast right. Keep an eye on items that have consistent differences between sales and forecast:
- Increase the forecast where your sales have consistently exceeded your forecast.
- Decrease the forecast where your sales have consistently been lower than your forecast.
Aligning your sales and forecast more closely, lessens the risk of costly stock-outs or generating excess inventory.
Adjust the forecast for new or lost customers, as soon as you become aware of the change. Use the computer forecast but:
- Subtract a lost customer’s monthly demand from the computer forecast.
- Add in the new customer’s expected monthly demand to the computer forecast.
- New stock items have no sales history, so these should be manually forecast for the first few months. However, check to see that the “new” item is not: A replacement for a “like” product, where a cheaper or better quality product has been sourced and will be sold instead of the old product. Linking the “new” item to the “old” item in a supersession chain will result in the sales history of the old item being used to generate the forecast for the new item.
Adjust your forecasts to include possible promotional demand on top of the regular sales demand.
Report on forecasting performance and make sure your measure also distills the bias between over and under forecasting. Make sure to measure the difference between the system generated forecast and the manually adjusted one to determine whether the manual intervention improved the result.
Finally, conduct a sanity check at a macro level. After making changes to the individual item forecasts on an exception basis, review overall sales to forecast to ensure that the overall growth is not too extreme or too conservative. Most forecasting solutions will enable macro forecast adjustments to be made if needed.
Once the monthly forecast review and endorsement has been completed, the focus switches to a weekly review aimed at highlighting exceptions. Here, any severe variations between sales and forecasts highlight potential issues with the forecast on individual items. Reviewing these alerts enables prompt response to possible changes in demand.
Review forecasts for the top 5-10 sales versus forecast exceptions:
- Where you are selling more than the forecast, increase the forecast.
- Where you are selling less than the forecast, reduce the forecast.
- Be mindful that you may be selling less due to stock-outs.
If you see a trend emerging, adjust the forecast immediately before it results in stock-outs or excess inventory.
A reminder: why do you want the best possible demand forecast?
The forecast is vital because it is your best guess as to your future sales expectations and is used to determine the optimal inventory levels that drive replenishment. Additionally, the forecast is used to determine whether the item is stocked-out, potentially going to stock out, is in excess, or you have too much on order. Without an accurate prediction of what you are likely to sell, none of these predictions would be possible, believable, or actionable.
The more fitting the forecast, the more accurate your purchasing and planning will be.