About the Blog

The FICO Labs blog focuses on innovation, technology and the role of analytic sciences in today’s business world. We’ll share our perspectives, vision, successes and challenges in the areas of predictive analytics, Big Data, decision optimization, cloud computing and Big Marketing.

« Covering the Spread | Main | You Are What You Measure »

Location-Based Marketing Using GPS Data

Orbiting Satellite
By Shafi Rahman and Amit Sowani

Collecting GPS data is becoming quite pervasive. Using the knowledge of where a customer goes, which path she travels, and how much time she spends at various locations can improve the quality of customer interactions and types of marketing offers, and increase the likelihood that she’ll redeem an offer.

GPS data is a time-series of an individual’s position information in terms of latitude and longitude. This data provides a wealth of hidden predictive information about your customers’ activity that could be used to improve marketing decisions. Working with clients, we’ve been able to gain insights on how best to leverage GPS data, including the importance of:

  • Measuring the duration of travel vs. the duration of inactivity. We found that these were quite predictive of customers’ future behavior. We can also gain insight into customers’ life-stage and lifestyle by analyzing the time of the day they had maximum movement.
  • Identifying the anchor points of each customer. Anchor points are places where people spend a lot of their quality time, e.g., home, office and school. Once movement with respect to these anchor points is identified, it can be leveraged to influence purchase behavior.
  • Determining the interaction between the path taken by the customer between anchor points and the places of interest (POI). This provides us with a powerful mechanism to influence a person’s redemption behavior. For instance, if there is a Pizza Hut on the way from a person’s office to home, she is more likely to redeem a Pizza Hut offer than a person who does not pass by Pizza Hut.
  • Measuring the maximum distance that a person travels from the anchor points and how often. Folks travelling quite far between anchor points are distinctly different than those traveling shorter distances.
  • Inferring interests and hobbies by the types of POIs frequented. For example: If the most visited location was a movie multiplex, then we could easily infer that the customer was interested in movies.

GPS data also presents challenges. The manner in which GPS data is collected today can be highly fragmented, with large missing values and high variability in the frequency and spread of the location points. For instance, for one client, we saw several cases where a person appeared to have travelled more than 500 miles in a time interval of 5-10 minutes. This was attributed to how the GPS data was being collected using mobile phones. Those interested in working with GPS data should watch out for such issues. More reliable mechanisms for collecting GPS data would yield better value.

It is important to note that in many countries, the use of GPS data is bound by strict privacy laws, requiring an explicit opt-in from customers. So how can you responsibly collect this data (one of the few information sources available on customer activities outside merchant stores)? You have to give to get. Give your customers a reason to opt in; it could mean giving them more relevant offers and discounts as a consequence of this opt-in, or providing a service resulting in a richer user experience.

First time on the FICO Labs Blog?
Subscribe to the FICO Labs Blog Feed or check out some other recent posts:

Comments

Post a comment

If you have a TypeKey or TypePad account, please Sign In.

Search Site


FICOLabsBlog.fico.com

Subscribe

Enter your Email

Preview | Powered by FeedBlitz