The retail world is abuzz with stories of omni-channel transformations. The promise of ‘sell anywhere, fulfill anywhere’ capabilities are driving retailers to adopt innovative approaches to increasing customer loyalty. The concept of ‘save the sale’ is a common component of these approaches — with retailers seeking to allay lost sales from out-of-stocks with solutions that range from simply providing store associates ecommerce access, to intelligent distributed order management algorithms that seek out the optimal source for a customer order. But, regardless of the path taken to enable ‘save the sale’ capabilities, every retailer ultimately asks the same fundamental question: “How can I make this as profitable as possible?”
Answering this question requires figuring out how you arrived at the problem to begin with. To find yourself in a ‘save the sale’ situation, you had to face opposing inventory challenges. On the one hand, you were concerned about having the proper in-stock position to avoid unnecessary stock-outs. At the same time, you wanted to maximize your inventory turns by setting the optimal targets to avoid overstocks.
The solution for profitably saving the sale lies in optimum safety stock and replenishment buying practices. Your in-stock position should be protected by your safety stock which defends against uncertainty in supply and demand. Your inventory turn goal should be the result of smarter replenishment buying.
Let’s examine those components more closely.
Safety Stock Optimization
Your inventory optimization system should have a safety stock optimization method that enables you to safely simulate your safety stock investment before applying it in production using:
- Multiple objectives
- Multiple performance measurements
- SKU families or SKU-by-SKU analysis
From my empirical observation as a practitioner in retail, this is still basically as true today as it was 20 years ago. The problem many retailers have (but do not realize it) is they have too much money invested in safety stock. Even assuming one is using an already somewhat optimized replenishment solution, our experience shows us that even then safety stock can be further reduced without jeopardizing the desired in-stock level at the store. Remember, safety stock is a defense against uncertainty in supply and demand. It should be demand-driven and take into account variance in supply and demand. Sometimes there are merchandising reasons like presentation stock that might need to override a statistically generated safety stock value, but this doesn’t need to be a blanket policy. Instead, it is better to have an application that can generate a suggestion of the most profitable safety stock investment and be intelligent enough to incorporate presentation stock.
In the example below, in the second column from the left, we see a simple stratification of SKU unit and dollar volume A – F. The safety stock factor is weighted depending on to which category (A, B, C, etc.) a given SKU belongs. This is a fairly intuitive way to think about safety stock. It makes sense you would want your ‘A’ SKUs in-stock guarantee to be better than your ‘F’ SKUs. Note that in this example with a safety stock investment of $90,827, we would have, on average, about 98.5% in-stock on all the SKUs in the sample. Using a more sophisticated method of in-stock simulation called ‘Minimize Lost Sales,’ we can optimize to the overall 98.5% in-stock rate and still achieve that overall in-stock but on a smaller safety stock investment of $86,014.
You should not settle for any replenishment solution that cannot enable you to simulate multiple safety stock policies using side-by-side comparison before you find the right combination of service and investment for your enterprise. Even in an otherwise well-run enterprise, this is an area where significant inventory savings can be found while maintaining the desired in-stock position.
Smarter Replenishment Buying
Part of smarter buying is knowing how much time supply you need for each SKU in the enterprise. Retail stores and other sales channels do not (or at least they should not) run in a vacuum. So, in other words, apart from the safety stock component that we’ve already discussed, what are the other components we need to consider when deciding how much to order for a given SKU? Lead time is another standard component. But the other component where you can have the biggest impact on the inventory investment is the average stock you carry for each SKU that should be enough to last from one receipt until the next. This is your cycle (for order cycle) stock. Of course we have our safety stock to bail us out in case something unusual happens, but in a perfect world cycle stock would last from one receipt to the next. Back to the point that store replenishment should not exist in a vacuum.
The store inventory investment should be managed and considered by the replenishment solution along with inventory at the distribution center. To understand ‘how often’ to order, your replenishment system needs to be able to simulate the cost of ordering every day to only ordering once per year and every combination in between. This is accomplished by balancing the cost of ordering versus the cost of carrying goods. Then, based on the lowest intersection of those two costs suggests the most profitable time supply, or said another way, on average how often to order. You might think, “We control the delivery to the stores and for other operational reasons we have to ship X times per week to each store anyway.” That may indeed be your operational reality depending on your business model, but if you are shipping from your own distribution center then you should have a system that can seamlessly manage the holistic inventory investment for the DC and the stores (not a siloed model where there is a separate ordering system for the DC and a separate fulfillment or allocation system to the stores). And in that case you would still need to simulate the economy of best time supply at the distribution center. And of course if your business model has you doing DSD, then this principle still applies.
Save the Sale…and Your Margins
No matter whether your store replenishment is single- or multi-echelon, your replenishment and fulfillment solution should manage the inventory investment for the enterprise in a holistic, non-sequential manner. Additionally, these processes need to take into account profitability. It is not a sustainable model if you are spending a dollar to make a dime. In fact, all too often we hear from retailers that, in their rush to ‘save the sale,’ margin erosion is the unwanted side effect. Implementing the proper inventory optimization solution should enable you to lower your overall inventory investment while maintaining or improving your service. In other words, hitting your desired instock levels and achieving your inventory turns in the most profitable way.
Rod Daugherty is senior director of product strategy for the Demand Forecasting and Inventory Optimization solutions at Manhattan Associates. For the last 19 years, Rod has been a consultant, designer and product director for multiple replenishment software companies including E3, JDA, Evant and Manhattan Associates. Prior to his career in Inventory Optimization software, Rod held various positions at a 1000-unit drugstore retail chain for 13 years (Thrifty-PayLess, now part of RiteAid), first in store operations management and later in corporate office merchandising.