Assortment optimization is a major focus area for retailers at the moment. Retailers need more accurate SKU rationalization, a better understanding of what to buy in the future, and localized assortments that drive sales. Getting this right is crucial in the market today. There are a few traditional ways that retailers plan out their assortments. They either 1) think through their assortment from a business process standpoint, 2) look at loyalty and shopper data to determine assortment, or 3) think of a brand-new method to determine assortment.
The problem with using what was an efficient business process to determine assortment is that it doesn’t necessarily lead to a profitable assortment. Shopper data is helpful, but the retailer needs to have vast amounts of customer data and understand what parts of the data are meaningful to good assortment. Although many retailers have volumes of data, the challenge is finding the right data and interpreting it correctly to make better decisions. Retailers looking for more innovative methods to improve supply chain generally fit the best with the solutions offered at 4R.
Our team has analyzed this common inventory problem and we de-aggregated demand down to the attribute level. This means our solutions can show what assortments are needed by location and by attribute for each store, allowing retailers to view inventory at a meta-SKU level. This provides kind of an X-ray view on shopper activity and helps fill in what each store needs based on actual demand and the market. This analysis shows how localization is built on the foundation of assortment optimization.
Without some level of localization, you certainly would have an assortment, but it would not be optimized. Every category needs some level of localization whether it be by cluster or down to a store-specific assortment. Knowing the right level of granularity is the key to offering a selection that customers want and will move fast enough to justify the inventory.
To many retailers, it may seem more cost-effective and efficient to order the same 200 SKUs for every store in a chain. The assortment is all about the what and not how much. What inventory should be kept at a DC or at a store is a matter of the forecast and optimized inventory. Efficiency is driven by demand, margin, and inventory carrying costs in order to drive the maximum profit. An efficient “buy” does not always yield the most efficient “profit.”
Some retailers may have very effective promotions, and therefore don’t think localization across the entire organization is necessary. However, localization at an attribute level drives demand by matching consumer wants. Promotion is done to drive demand when there is price sensitivity but does not necessarily increase profits. The machine learning attribute demand models tell you nuances in your customer base and can maximize sales without unnecessary promotions or markdowns.
Localization also helps increase profits by helping retailers select the best assortments. Most categories increase in sales when they are a very granular level of localization. Other categories do not need as many “versions” of assortments. Modeling out the demand increase at each level of granularity shows you a “build curve” for each category and where the optimal level of localization is achieved.
Even with one master assortment, there is always room for improved optimization that can be achieved using the attribute demand models. Demographics, income levels, psychographics, and other factors all play into individual stores’ consumer groupings. There is always a level where localization can increase demand and decrease demand cannibalization. However, without the right tools in place, this level of optimization is impossible to find. Traditional methods will fall short or overcompensate. Assortments powered by machine learning, AI, and advanced analytics make it easy for retailers to maximize profits.
Kevin Stadler | President & CEO
Kevin Stadler is the President and CEO of 4R Systems. He is responsible for the continued strategic expansion and overall operations of 4R. Stadler has over 25 years of experience in retail technology, product strategy, sales, and software design. Prior to joining 4R…
Read more about Kevin here.