Uncovering Behavioral Segments by Applying Unsupervised Learning to Location Data

Abstract: 

Location data is a powerful tool. The places people go reflect who they are and what they care about – especially during the holiday season, when shopping is at a high. During the holiday season, shoppers often deviate from their normal habits, shopping more frequently and engaging in larger purchases. Therefore, to help marketers understand and best meet holiday shoppers’ needs, Foursquare applied unsupervised learning methods to location data to derive meaningful segments of individuals based on their demographics and shopping behaviors. Rather than invest in reaching more general segments like Moms or Millennials, marketers are now able to focus their efforts by targeting these data-driven segments during the 2022 holiday season.

Dimensionality reduction methods such as Principal Component Analysis (PCA), combined with clustering methods such as KMeans, can isolate which features describe the most variance between users, then use those features to group like users together in an unsupervised manner. From there, performing analytics on the segments helps to further understand their affinities and shopping behaviors. This information empowers marketers to determine which segments present the largest opportunity during the holiday shopping season, and what strategies to use to best target them.

To demonstrate this approach even further, Ali will walk through recent research conducted by Foursquare’s data science team. The team analyzed 2021 holiday shopping behavior and was able to identify six key data-driven audiences, including:

Expert Shoppers
Expert shoppers are an undeniably important audience to pay attention to during the holidays. These shoppers are frequenting retailers more often than any other audience, and also spending the most time in stores during the holiday season.

Weekend Drivers
Weekend drivers tend to be mature shoppers — 71% of this audience is over the age of 45. They’re also mostly male (59%). As their name suggests, they prefer to shop on weekends — 43% of their holiday shopping occurred on weekends between Black Friday and Christmas last year.

Suburban Spenders
Suburban spenders are the most affluent shoppers of the bunch — they’re 72% more likely to make over $100K annual income relative to the total U.S. population. And as their name insinuates, they’re also 16% more likely to live in suburban areas.

Urban Light Shoppers
Over 50% of this audience is between ages 18-34, and most are living in big cities — compared to total U.S., they’re 107% more likely to live in urban areas, 3.2X more likely to frequent metro stations, and 1.6x more likely to frequent laundry services. They’re also more likely to shop later in the day — over 60% of holiday shopping happens after 2PM.

Weekday Browsers
Weekdays (Monday - Friday) account for 90% of total holiday retail traffic for this group. They’re also more likely to shop earlier in the day (over 50% of their holiday retail visits occur before 2PM). Weekday shoppers are visiting the fewest retailers.

Rural Value Shoppers
Rural value shoppers are 60% more likely to live in rural areas compared to the average U.S. consumer. They’re also 2X more likely to frequent discount stores.

These six new and unique data-driven audience segments are available to activate everywhere marketers buy media this holiday season.

Background Knowledge:

Fairly Familiar

Bio: 

Ali Rossi is a Data Science Tech Lead at Foursquare, working closely with their first-party foot traffic panel to deliver insights against a broad range of client business questions. She is passionate about consumer behavioral data, with experience building consumer panels, researching normalization methodologies, and developing methods to derive actionable insights. Previously, she worked in product management at Foursquare, Amazon and Nielsen, mainly focused on building analytics products using consumer-sourced data. She studied chemistry and mathematics at the University of Connecticut and is currently pursuing a Master of Science in computer science at the Georgia Institute of Technology.

Open Data Science

 

 

 

Open Data Science
One Broadway
Cambridge, MA 02142
info@odsc.com

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from - Youtube
Vimeo
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google