How To Make Half-a-million Inventory Decisions A Week
By Steve Searfoss

A shoe store with 400 different styles in 10 sizes and 3 colors has a total of 12,000 possible items to stock in a store. If we multiply this number by 40 different stores, then we are talking about 480,000 possible items in inventory for this particular chain. This means that the management of this shoe chain  must make half a million supply policy decisions every three months (if only by omission) and take half a million supply actions every week. A supply policy decision is asking, “How many pairs of Suede Boots 2214 do I want in black size 4 in store 37?” A supply action on the other hand, is, “We decided to have two pairs of our 4829 Boots tan color size 5.5 in store 19. Do we still have two there? Or Do I need to send them one or two more pairs?”

Sound complicated? It is. There is no way half a million decisions can be taken correctly and efficiently without the help of sophisticated software like XpertMartTM that automates the process by monitoring sales on a daily basis and suggesting purchasing and distribution decisions.  If they are to reap all of the benefits software affords, retailers must harness the information such a system provides across their entire distribution system, changing the way they purchase merchandise, receive shipments in their warehouse, transfer out to their stores,  move items between stores and ultimately, purchase again starting the whole process over. Systems and Operations must work hand-in-hand.

A retailer’s purchasing and distribution system should be set up to bring supply as close as possible to demand, thereby dramatically reducing unnecessary inventory and minimizing lost sales. Carrying unnecessary inventory creates financing costs and forces a retailer to sell at markdowns, eroding margins in both cases. The flip side is not having the correct item in stock and therefore failing to satisfy a customer. This unsatisfied customer is not one of the millions of potential customers somewhere out there, but one of the select and valuable few who made it into your store, liked one of the styles you have, was willing to pay the price you ask, but then… you didn’t have the correct color and size. To use a baseball analogy, this is like striking out with the bases loaded and two outs.

Those retailers whose distribution systems are set up to understand and respond to their demand will be the winners in the marketplace. The way to read demand is through Minimums, which set a minimum stock level for each item, in each store, based on actual historic sales. The Minimum serves as an equilibrium point: stock levels above the Minimum mean you are overstocked and keeping capital unnecessarily tied up in inventory that will not sell; stock levels below the Minimum mean that you are understocked and risk losing sales. So a retailer’s supply system must be calibrated to keep the stock levels (item by item) as close to this equilibrium level as possible. Just to keep things interesting, this equilibrium level is a moving target as customer’s tastes are constantly changing.

In an ideal distribution system, Minimums should be used to control distribution of merchandise from a central warehouse to the stores. Therefore, the actual Minimum depends on the time it takes to move an item from a warehouse to a given store. Maximums, on the other hand, should be used to control the purchasing of merchandise from vendors and its shipment to the central warehouse.

Suppose it takes one week for your warehouse to process a transfer order and send an item out to a store. To keep the example simple, we would want our Minimum to be the amount of items sold during that seven day period. Now suppose that for this particular style, our vendors need four weeks to re-supply our warehouse in which case, we would want our Minimum to be equal to the amount of items sold during that seven day period multiplied by four (or even better, the amount of items we have sold in the last 28 days). 

 To continue with our example, let’s suppose you last week a certain store sold three white, size 4 “Tommy” sneakers. This store would order 3 more white, size 4 Tommys from the central warehouse. At the same

time, the warehouse would order 21 white, size 4 Tommys from the vendor. If the following week the same store sold 4 white, size 4 Tommys, it would order 4 more pairs from the warehouse, which would, in turn, order 28 pairs from the vendor. We can imagine a line of imaginary trucks full of Tommy sneakers each week waiting their turn to deliver the shoes to our hypothetical store. 

In a sense, the Minimum represents a safety margin to cover flukes in sales or breakdowns in distribution. At the most basic level, you want your Minimum to be equal to a certain amount of “days of sales” and for that you need two parameters: a sales period and a multiplication factor. As we’ve mentioned, the number of days of sales you want to keep in stock depends on the time it takes for your warehouse to distribute merchandise to your stores, plus a safety buffer. Suppose your warehouse takes 10 days to supply your stores and you want to keep a safety margin of 5 days of sales, then you would want your Minimum to be equal to 15 days of sales. In this case, you could take August 1 – August 10 as your sales period and 1.5 as your multiplication factor. This means that the Minimum would be the number of items you sold between August 1st and August 30th  multiplied by 1.5. If you sold 6 items during this period, your minimum would be 9. 

To these two basic parameters, XpertMartTM lets you add two more: a lower limit and an upper limit. We use the lower limit to set an absolute minimum value, even if the product of the two basic parameters is lower. We do this, for example, if we believe that sales during the period of analysis were abnormally low for some reason (such as bad weather) or believe that there is inherent value in having a certain amount of items in stock, for example, having at least one in each size of a style to always be able to tell a customer: “Yes, we have it.” (In some market segments, presenting an image of abundance is a virtue).

Similarly, the upper limit gives us an absolute maximum value, even if the product of our two parameters is higher. We use the upper limit for opposite reasons: if we believe that sales during the period of analysis were abnormally high (a holiday perhaps) or if we are so certain that we can restock stores quickly (either from our warehouse or through our vendors) that there is no need to keep such high levels of stock or if there is a lack of storage space in the store. 

You should take great care in setting these parameters. Much of the precision of Minimums depends on the human factor. When choosing a historic sales period to use as a parameter, make sure it is not atypical, such as Christmas or Mother’s Day when sales are unusually high. Using December’s sales to set January’s Minimums is sure to lead to overstocking. If your business is highly seasonal, for example, selling T-shirts and swim-wear to tourists, it might be more appropriate to use the same sales period from the prior year (i.e. June – August) than the three previous months (i.e. March – May) which are part of the low season. Use real data whenever possible. If your goal is to have 60 days of sales in inventory, it is far better to take a two month sales period and multiply by a factor of one, than to take one month of sales and multiply by a factor of two. The resulting Minimums will be far more accurate the more you rely on actual sales data. 

Minimums will be most effective if you try to differentiate parameters rather than use a uniform set of parameters for your entire stock. You may want 15 days of sales of stock in your stores, but 25 days of sales for a particularly popular style if you anticipate that sales will continue to grow. Using our XpertQueryTM tool you can filter the parameters such that they are assigned only to certain brands, styles, stores, vendors or even sizes and colors. Finally, be sure to periodically re-calibrate your Minimums. As you start using Minimums to optimize your inventory, your sales will undoubtedly rise--in which case your Minimums should increase as well.

About The Author:Steve Searfoss has been helping retailers optimize their inventory and grow their sales for over five years. He is CEO of XpertMart which makes POS software