The Missing Link In “Going Local”

  • Written by  Marek Polonski, APT

0aaMarek Polonski APT“Going Local” has been a key ambition for retailers for years now. Recently, retailers’ desire to localize their offerings has manifested in new store formats, with many big box players focusing on their smaller footprint neighborhood stores. Others are differentiating in more nuanced ways, slightly altering merchandise allocation in each store or group of stores to meet the needs of the local shopper. While localization has been front-and-center in the industry for a decade now, retailers still have a significant opportunity to improve their strategies to better meet local market demands.

There are many components of a successful localization strategy, including matching marketing and pricing efforts with the competitive environment as well as demographic and psychographic factors. However, one often overlooked facet of an effective localization strategy is a connection to the space planning process. Understanding how to tailor assortments for local market demands can ultimately lead to significant profit improvement and a better customer shopping experience. But how can retailers achieve these objectives?

One critical component is understanding the best space allocation for every category and department. However, despite the desire to “go local,” many retailers have under-invested in their space allocation capabilities. Even with vast improvements most retailers have made in other analytic capabilities, most organizations still rely on business intuition or very basic analytics to make space tradeoffs. These approaches limit their ability to understand the true relationship between space changes and sales. The good news? Taking space planning to the next level of analytic rigor is attainable and profitable.

Why are sophisticated analytics necessary to improve space allocation and ultimately offer a more localized experience? The challenge lies in understanding the most productive use of every foot of space. Every space decision is ultimately a tradeoff between two or more categories: when one category gains space another one must lose space.

The default approach to making these tradeoffs has generally fallen into one of two buckets: intuition (I think this product will sell more if we give it more space) or rudimentary analytics, such as using average productivity (sales per foot) to determine space allocation. While this second approach is a first step in leveraging space and sales data, it does not account for the incremental impact of space changes and the diminishing returns every category has when adding space. Said another way, how much sales or profit will an added foot of space generate?

Consider a retailer that has just installed a new fixture with four additional linear feet of space, which it can either allocate to Salty Snacks or Candy. On average, Candy generates $400 per square foot and Snacks generate $250, suggesting that Candy should receive more space. However, analyzing performance across similar stores with different space allocations for these categories may show that adding one additional foot of Candy only generates $150, whereas adding one additional foot of Snacks generates $200. These insights would change the retailer’s merchandise decision and generate an immediate profit opportunity with the same investment level.

While understanding the optimal space allocation across the chain is important, to truly localize assortments retailers must take this rigorous approach with each store or store cluster. After all, allocating additional space to swimwear in Florida is likely more profitable than in Kansas City. To get started, retailers should analyze all of the factors that might be driving higher or lower category productivity, such as box size, surrounding store demographics, competitive presence and weather. Combining these factors with operational considerations, retailers can then create clusters of stores to analyze together.

In one example, a retailer used advanced space planning analytics to identify an opportunity to generate a 5%-7% increase in sales by making small shifts in current space allocation. By understanding the marginal returns of each category, they could better understand how to merchandise their stores to meet local customer demand.

The retailer took this approach a step further by identifying which store and local market characteristics were driving higher space productivity for each category. After identifying and quantifying the statistical influence of these characteristics, the retailer created clusters that behaved similarly. By transforming their space allocation approach from one that one was uniform to one that varied by store cluster, the retailer was able to double the potential impact of space changes in target categories.

“Going Local” is a strategy that has significant potential to unlock financial value and improve the customer shopping experience. Realizing this value requires commitment from senior leadership to think differently about how they shape assortments to align with customer demand. Executives can leverage space optimization analytics to reallocate space across stores, and cluster stores by specific market drivers of demand for their categories. With the right analytics in place, retailers can take another step in the right direction of tailoring their value proposition to local markets.


Marek Polonski is a Senior Vice President at APT, and has extensive experience applying Test & Learn principles across a broad variety of industries and functional topics. He has advised Fortune 500 corporations on enhancing the profitability of capital investments, pricing optimization, labor allocation and training, merchandising initiatives, advertising, and operational optimization. He has also helped retailers institutionalize the rigor of using analytics to drive business decisions. Polonski’s experience spans banking, retailing, manufacturing, restaurant, retirement services, and other industries. He holds a BS in Computer Science and BS in Economics from Massachusetts Institute of Technology.

back to top