Communications of the ACM
Foursquare Check-Ins Tell Stores Where to Set Up Shop
Analysis of 35 million publicly available Foursquare check-ins from 925,000 users in New York over a six-month period provided insights into the best locations for new stores.
Credit: Foursquare
University of Cambridge researchers recently conducted a study on the use of social media to help retailers find the best locations for new stores.
The team analyzed 35 million publicly available Foursquare check-ins from 925,000 users in New York over a six-month period, concentrating on three specific chains. Stores were rated based on the number of check-ins received, and multiple check-ins by the same user were modeled into a personal movement pattern to gauge overall foot traffic, how frequently people travel long distances to visit an area, and other measures.
The team then assessed the ability of each factor to forecast a store's success. Check-in data for two-thirds of the stores was fed into an algorithm, which then predicted which of the remaining third of the stores would be the most successful.
Although each factor alone proved highly predictive of a store's popularity, a system incorporating all of the factors was the most effective, correctly forecasting the top 10 percent of locations 70 percent of the time.
From New Scientist
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found