Using Footfall Analytics to Improve Store Performance

a view of a large building

For retailers, COVID-19 has had consequences far more serious than a rainstorm, or even a blizzard. Yet both the weather and coronavirus have certain characteristics in common.

Like the weather, the effect of the pandemic is variable and yet, predictable. Surges in infection rates vary city by city, and neighborhood by neighborhood. Unlike the weather, however, COVID-19 can even alter customer shopping patterns street by street.

In some small pockets of communities, foot traffic may change dramatically on one street yet two streets over, analytics will show not much change at all. In urban areas, the coronavirus can actually cause shopper traffic to increase, as people who normally commute are now working from home. Meaning, during the weekday, they are going for coffee, making drug store visits or walking their dog, all in their local neighborhoods. Retailers can even see changes within their stores; for example, COVID-19 will affect the likelihood that someone will linger in front of a product display.

Advanced spatiotemporal analytics — the study of movement over time in a particular geography — is answering questions like these for retailers, in ways unheard of just a few years ago. What’s more, the discipline is enabling merchants to make adjustments that can improve performance at the store level, at regional levels, even on an aisle-by-aisle basis — and do it in near real-time.

Thanks to the latest CPU and GPU processing breakthroughs, accelerated data analytics is able to process billions of rows of data in milliseconds, without downsampling or data aggregation. Using location intelligence technology, these results can be plotted geospatially in map form for real-time analysis by both data scientists and non-technical users.

Spatiotemporal analytics answers numerous questions that can bring insight to retail decision-making. It identifies patterns and anomalies in foot traffic. For example, if shoppers are spending less overall time in front of particular store displays due to the pandemic, are there specific instances when that isn’t the case? Are there patterns among customers, or groups of customers, that indicate when they are less likely to observe COVID-19 safety concerns?

Data science is able to not only identify those patterns and anomalies in milliseconds, but over time can actually predict them using artificial intelligence. Spatiotemporal analytics can also compare in-store patterns to micro-level data about what’s concurrently happening in the broader community, then apply those learnings to store-level economics and patterns.

Various datasets can then be combined for targeted promotions, loyalty programs and the like. Shopper movement data can be leveraged against customer purchase history to optimize real-time offers, promotions and product information. Such notifications, delivered via the retailer’s app or by other communication channels, can be timed for when the shopper is in the store vicinity or even as they move in-store through certain departments.

Even more uses are possible when merchant data is cross-referenced against any of the thousands of public datasets that are freely available. Census data, traffic and transportation updates, point-of-interest information, macroeconomic data and of course public health data can be utilized to bring new insight to in-store and out-of-store behaviors.

From advertising to merchandising, in-store services to store placement, footfall analytics can inform critical retail decisions. As a result of COVID-19, the survival of physical retail is all about how people behave at the local level, and spatiotemporal analytics can drive the kind of fresh innovation retailers need to more effectively reach consumers.

It’s true we can’t do anything about the variability of the weather. But foot traffic analytics can bring new clarity to store decision-making — and perhaps, improve the chances of a sunnier retail forecast in the future.

Nohyun is the VP of the solution engineering practice globally for OmniSci and brings extensive experience as a technologist, strategic executive and board advisor spanning 20 years in the data, analytics and high-growth technology space.

More on Analytics