During my 20 years in the retail industry, I’ve witnessed retailers transition from paper and calculators to spreadsheets, to on-prem software, to cloud based software and now to artificial intelligence-enabled automation. I’ve been fortunate to see some of the best in the industry, like Nike, Under Armour, and Sephora, elevate their processes to fuel impressive growth.
I’ve heard a lot of myths about how things must be done or why they can’t be improved across the spectrum of retail use cases. But one area that has always been close to my heart is how technology can help retailers more accurately allocate their inventory to both physical and virtual locations. In this article, I explain the top four misconceptions I’ve heard and how to overcome them.
Misconception 1: Allocation is About Where Inventory Goes
Many retailers see allocation as the final, uninteresting step after all the hard work of designing, planning and buying a great new assortment. All too often they simply set minimums and maximums (min/max levels) per store, distribute the inventory and hope for the best.
Yet, modern allocation is about allocating demand, not inventory. In this key distinction, a retailer builds their optimization engine to place inventory to meet demand in the most optimal way. As that engine reads performance inputs, it becomes smarter, more efficient and more accurate.
A demand driven strategy optimizes the supply chain from top to bottom to maximize availability without additional inventory investment resulting in better sell-through.
Misconception 2: Historic sales are the only factor to consider when allocating.
Flagship stores, rightly, deserve significant attention from the allocation team to ensure the brand image of a retailer is maintained and stock-outs avoided. However, this focus can often divert inventory from smaller stores creating a self fulfilling prophecy in the smaller stores of poor performance and therefore diminishing allocation and ultimately closure, all due to the fact that they didn’t receive the right inventory.
Retailers should consider implementing presentation standards across stores, this standard ensures that each store has the opportunity to sell, especially lower performing stores that previously have been starved of inventory leading to future reduction in allocation quantities if based purely on historic sales.
After initial allocation and an opportunity to sell, presentation priorities can be adapted to prioritize demand and ensure that the remaining inventory is pulled to the store most likely to sell it. Only after true demand is understood should we put more emphasis on historic sales in the allocation strategy.
This change ensures stores have an opportunity to perform, growing the brand, sales and ultimately profitability
Misconception 3: Some SKUs are Impossible to Predict
Many retailers throw up their hands on attempting to plan for slow-moving, intermittent SKUs. While forecasting these items comes with inherent difficulties, demand projections are incredibly important in optimizing allocations. The ability to project an uptick in sales due to a promotion or holiday can ensure that stores have enough stock to meet customer needs, which traditional min/max strategies struggle to do.
Various machine learning (ML) and artificial intelligence (AI) algorithms can be used to forecast slow moving items with a fair degree of accuracy. In particular, AI can help with demand forecasting by taking into account a wider range of factors than traditional forecasting methods, such as seasonality, trends in consumer behavior, and changes in external factors like the economy or weather patterns.
Even a small improvement in demand accuracy can ensure excess inventory is reduced, ultimately resulting in lower markdowns and increased profit.
Misconception 4: I Need a Whole Team of Allocators
Growth can sometimes feel like a curse. One successful store leads to another and another and another and before you know it, stores are running out of best sellers and customer loyalty begins to suffer when they can’t find their size or color. It can seem like the only solution is to throw a team of humans at the problem.
However, I’ve seen much better success when a single allocator simply has the right tool for the job. With an advanced allocation solution, this one resource has more power to move the needle on in-stock rates and full price sell-through than a team with spreadsheets. The power of a demand driven allocation strategy solution is that once the strategy is set, the system manages all the changes over time and highlights exceptions while automating a majority of the allocation decisions.
Not only are the decisions better, resulting in improved customer loyalty, but the business saves on employee expenses as well.
Why Successful Retailers are Focusing on Allocation Optimization
Successful retailers refuse to let these misconceptions cost them with excess end-of-season inventory in the wrong place and costly transfers eating into their margin. Instead, successful retailers are embracing allocation as a way of closing the loop between buying and selling to improve margins and free up cash flow for further investment into the assortment.
With two decades of experience in retail, Peter Leith has collaborated with some of the most iconic brands in retail (Levi Strauss, Williams Sonoma, Chanel) to improve their merchandising, buying and allocation processes. As the leader of Toolio‘s product team, he aligns the needs of our customers to the technology we build.
