If you’ve worked with our Retail Gap and Retail Potential data over the years, you know these are two of the core tools for understanding retail demand, supply, and opportunity. At their simplest, they answer a very practical question: Are consumers spending money here—or somewhere else?
But as with most things in data, “simple” only works if the underlying methodology keeps up with how the world operates. And retail, especially over the past few years, has changed quickly.
With our upcoming release, Spatial Data Scientist Alex Burdsall took a fresh look at how these datasets are built—from the ground up—and made some meaningful updates that better reflect today’s retail landscape.
Let’s start with one of the biggest improvements: how we connect what people buy to where they buy it. Historically, this relied on older Census of Retail Trade merchandise line relationships. Those worked well for a long time, but the retail world has evolved. We’ve now rebuilt those relationships using the 2022 Economic Census product-industry files, which provide a much more current and detailed view of how merchandise lines map to store types. In practical terms, this means a more accurate translation of consumer demand into real-world retail categories.
Another major shift comes from how we handle e-commerce. Retail Gap has always reflected the difference between consumer demand (Retail Potential) and brick-and-mortar supply (Retail Sales). But today, a meaningful share of that “gap” isn’t leakage to the next town over—it’s leakage to the internet. With this release, we’ve updated national retail supply estimates using current e-commerce and omnichannel data, so the story you’re seeing is much closer to reality. In many categories, understanding why leakage exists is just as important as knowing that it exists.
We’ve also significantly expanded the level of detail in local retail sales estimates. By incorporating the latest inputs from sources like the 2022 Economic Census, the Annual Retail Trade Survey (ARTS), the Annual Integrated Economic Survey (AIES), and ongoing monthly retail and e-commerce data, we’re able to provide a much more granular and current picture of retail activity. These updates don’t just refine the numbers—they improve the reliability of the insights that come from them.
And speaking of granularity, one of the most impactful changes may be one you don’t immediately see: the data is now built at the census block level. Previously, much of this work was done at the block group level. Moving to blocks allows us to better capture hyper-local variation—those differences from one neighborhood (or even one side of the street) to another that often make or break a retail decision.
We’ve also expanded what “retail” means in this context. In addition to traditional categories, we’ve added Entertainment and Healthcare as merchandise lines, reflecting how consumers spend their time and money today. This includes new NAICS categories like motion picture theaters, spectator sports, veterinary services, medical offices, hospitals, and even childcare services. These aren’t fringe categories, they’re integral parts of how communities function and where dollars flow.
On the service side, we’ve broadened coverage in areas like automotive repair and maintenance as well, recognizing that consumer spending doesn’t stop at the checkout counter of a traditional retailer. The line between retail and service continues to blur, and the data now reflects that.
All of these updates build on the same core concept that has always defined Retail Gap: the relationship between demand and supply. Retail Potential still represents what consumers should be spending locally, while Retail Sales represents what is being captured. The difference—leakage or surplus—remains one of the most intuitive and powerful ways to identify opportunity.
What’s changed is how precisely we can measure it, and how confidently you can act on it.
For site selectors, this means a clearer picture of underserved markets. For retailers, it means better alignment between store formats and actual consumer behavior. For economic developers, it means a stronger foundation for recruitment strategies. And for anyone working with location data, it means fewer assumptions—and better answers.
As always, our goal is straightforward: good data, better decisions. This latest release is a big step forward in making sure the data reflects the world as it works today—not how it worked five or ten years ago.