Utilizing SafeGraph foot-traffic data and Led by Academics from the University of Wisconsin-Madison and the University of Virginia, Study Revamps the Traditional Huff Model to Improve Trade Area Analysis Accuracy
May 12, 2020, San Francisco (CA) – Academics from the University of Wisconsin-Madison and the University of Virginia, in collaboration with SafeGraph, a geospatial data company, released a comprehensive research study that developed an adjusted Huff model to predict the attractiveness of different types of businesses over time. This model is more accurate than the traditional Huff model in predicting the market share of different types of businesses over time, taking retail site selection and trade area analytics to a new level.
The traditional Huff model assesses a store’s relative “attractiveness,” using data on the store’s size and customer proximity to predict the likelihood of customers visiting that store over others in the area.
Convinced that the Huff Model could take our knowledge about foot traffic patterns a step further, researchers at the University of Wisconsin-Madison and the University of Virginia, in a paper entitled, “Calibrating the Dynamic Huff Model for Business Analysis Using Location Big Data,” explored how weaving ‘time-awareness’ into the traditional Huff Model could give retail trade analysis a whole new level of precision in its predictions. Powered by SafeGraph foot traffic data, this revised model is able to predict store patterns to supermarkets and department stores across the 10 largest US cities.
The paper improves the accuracy of the Huff model by incorporating “time awareness”. Using SafeGraph data, this improved model (the T-Huff Model) predicts times during the day where stores are most likely to see comparative spikes in foot traffic.
“We’re impressed with the richness and quality of SafeGraph’s business venue data,” said Professor Song Gao of the University of Wisconsin. “We see it particularly adds value to the adjusted Huff model we’ve developed. Using SafeGraph’s Points-of-Interest & business database, our team was able to calibrate a time-aware, dynamic Huff model through analyses of customer behavior and retail data.”
Key Findings from the study:
- The adjusted Huff model that incorporates temporal variability is more accurate than the traditional Huff model in predicting the market share of retail businesses over time.
- Demographic and socioeconomic factors like median household income have a significant impact on a customer’s choice of a particular store.
- Retail businesses can use these insights for setting business hours, allocating transportation resources, and improving accessibility during traffic peaks.
Below is a snapshot of the model, looking at the market share change of 5 Whole Foods locations in Los Angeles on Sunday at 3 PM. Green indicates areas where traditional trade area analysis would have under-reported a store’s market share.
SafeGraph is a data company that seeks to understand the physical world and power innovation through open access to geospatial data. SafeGraph has built the source of truth for physical places, covering business listing information, building footprints, and foot-traffic insights for 6+ Million Points-of-Interest in both the US and Canada. For more information visit the SafeGraph website, read our blog, and follow us on Twitter.
Pure Knot PR