IoT, AoT, and Retail Analytics Use Cases

Learn Data Science
Teradata Employee

Back Ground:

According to some recent research, shoppers still prefer to buy at a physical store rather than from an on-line retailer; furthermore studies have shown that people prefer to use their senses when making a purchase. However many customers go to a retail location to make a purchase and decide to buy on-line rather than at the store because of two main drivers, selection and price. Shoppers are going to use their senses at stores because they trust them when making a purchase, but buy on-line because the product was neither in-stock or a cheaper price was found on-line. Wouldn't it be great if brick and mortar retailers had a way to analyze foot traffic patterns much the same way that on-line retailers do with web click streams?


Wi-Fi enabled smart phones, with a market penetration of over 50% of the population, provide retailers a great opportunity to capture more sales from customers, improve the overall customer experience, and understand foot traffic patterns in their stores. As long as the Wi-Fi atena is enabled on the smart phone, phones will continually request access to the network even while the phone is sleeping. In many cases a phone will request access to a network 5 times a minute. Retail locations that have a Wi-Fi network along with strategically placed Web Access Points(WAP) or beacons, can begin to understand the paths people are taking while in the store, and find affinities between store departments. The captured location data can be streamed to Aster via Teradata Listener.


Once the data has landed in Aster, the real opportunity begins with the Analytics of Things.  Aster has a wealth  of built in functions such as nPath, JSON Parser, Sessionize, and CFilter that will accelerate the discovery process so that actionable insights can be found and acted upon. Using the aforementioned functions several discoveries can be quickly found:

  • the most common paths people take
  • department and product affinities
  • times days and departments that are the busiest

Use Cases:

The use cases that are available depend upon which type of shopper is on the network, whether they are known or unknown. Known shoppers are the ones that connect to the network and sign in with their social media account or some other means to identify them.

Known Shopper Use Cases:

  • Deliver Personalized Offers when they  enter the location or department
  • Enhanced Customer Loyalty through timely and appropriate communication
  • Customized Product Recommendations while in location
  • Mapping Web Click Streams with Retail Location Data to improve customized inventory
  • Customer Paths by Demographic Information

Unknown Shopper Use Cases:

  • Most Common Customer Paths
  • Product and Department Affinity
  • Optimized Staffing Levels
  • Optimized Floor Layout and End Caps

Privacy and Transparency:

Like any data capture and analytics program two vital components are necessary for customer support and buy-in, privacy and transparency. Being transparent with customers about foot traffic analytics, what data is being captured and stored, and what is being done with the data is crucial. Most customers are happy to give their data, but they want to understand what is being done with it, and is it secure.