Are people as anxious to get out of their homes and back into stores as Retailers are to re-open their doors? The retailers, suppliers and all the providers that service them hope so! This is unchartered territory. I remember the first time getting on an airplane after 9/11. It was scary but now waiting in TSA screening lines feels normal. And now, unfortunately, so does grocery shopping with a mask and gloves on.
That said, retailers and their supplier partners will have a huge task ahead to quickly gauge the consumer’s mood and watch for trends in stores. Tracking sales by location and optimizing inventory allocation will be the key- making sure the right stores have the right products. Looking at chainwide sales won’t work because only a subset of stores will be open. Planning and forecast teams will have to look at a variety of factors to figure out what to put into production to meet the new level of demand.
I recommend they start preparing now by retrieving door level history on key items and setting up a tracking plan to compare rates of sale from different time periods. I would start with last year for the same period, then the most recent 6 months, 3 months and the last 4 weeks. Use what makes the most sense for the product category and strength of item. Start simple with unit sales and on-hand inventory. Profitability and other KPI’s can be added later.
The key is to determine what period of time to base the forecast on. We calculate the rate of sale by taking the average units sold over the selected time period- including only those weeks where sales were greater than 0. For the last year’s history, be sure to adjust for any lost sales that may have occurred. Seasonality may also inflate or deflate the rate of sale depending on the time of year.
Now the comparative analysis can start. In my example, I want to see which direction the numbers are going. In some stores they are starting to return to last year’s levels, in others, they are still behind. To be safe, I have decided to base my new forecast on an average between last week and the last 4 weeks.
The next step is to determine how much inventory should be in each store based on the new forecast. Last year’s model allowed 12 weeks of supply, but now we will use 8. By multiplying my new forecast by the 8-week model I now know how much inventory is required. I can roll up all the store needs to see the total inventory requirement. If you are a supplier, it’s important to talk to your buyer to find out what their philosophy is on the model stock as it might have changed, and how much safety stock you should be maintaining.
Finally, I compare the inventory requirement to the wholesale inventory pipeline. In this example, the total need 329 units which is made up of the target store inventory of 188 units plus 6 weeks of safety stock. My current ATS and WIP is 254 so I need 75 units. Based on the lead time and minimums I can now make my buying decisions.
This exercise will have to be repeated weekly as more stores start to re-open and the rate of sale increases. Set up trigger dates for when you must make decisions and at least you will be using the best information you have at that time. Contact us if you need help completing this or want to start getting store-level data. We have the expertise and tools to run these scenarios on millions of SKU-store combinations quickly and affordably.