In George Orwell’s dystopian novel “1984,” the warning “Big Brother is watching” appears on posters throughout the fictional Oceania. And though it’s not for the purpose of wielding power, as it was in Orwell’s book, this phenomenon of tracking the whereabouts and actions of users or customers no longer is the far-fetched concept that it was when the book was published in 1949. With the emergence of artificial intelligence and machine-learning techniques, marketers and retailers increasingly are using location data in their marketing analytics and applications.
Location data come in degrees of granularity as a function of time. The most static form are about where customers live — their geographical locations, ZIP codes and neighborhood information. These data do not change much from day to day, or even month to month, and they tend to be pretty stable, with gradual changes over time.
{mosads}In combination with customer demographic data and market data, such data long have been used by marketers and the retail industry for geographical segmentation and for formulating strategies, such as those pertaining to store expansion and other location decisions.
Static location data are quite rich; they serve as proxy for variables that cannot be easily observed. For example, ZIP code data can serve as proxy for income and education variables of customers living in that locale, and they can capture social contagion, such as the adoption of online grocery delivery and other demonstration effects (e.g., people seeing what their neighbors buy and then copying them). Such location data are routinely collected and organized by many syndicated market research firms, such as Spectra, which provides trade-area demographic data.
The more dynamic form of location data is the one that has virtually exploded in volume and velocity in recent years with the advent of mobile phones and sensor technologies — that which reminds us of Orwell’s classic. These days, the concept of location data refers more to this ubiquitous form of customer- or user-location data as consumers move about within their homes, within stores, within and across cities, and all over the globe.
Depending on the country and local jurisdiction, privacy laws may limit the collection and transmission of such location data, since these actions may violate customers’ privacy, but many marketing applications using such data have emerged in recent years.
This can benefit customer and advertiser. For example, geotargeting at the individual level is done using mobile location data. Companies can provide coupons and special offers to users within a specified distance from specific stores. When you go to a mall and get close to Macy’s, you may get a special promotion from Macy’s prompting you to visit its store. Firms also can do geo-conquesting — that is, if data indicate that you are near a competitor’s store, the company then provides a promotional coupon for visiting its store, prompting you to choose it over the competition.
Search engines and smart agents, such as Siri, can personalize recommendations to users based on their locations. Yelp can suggest top restaurants close to your location at a given point in time. Many such applications on mobile devices are dependent on such location data to provide personalized content and service to users.
The most useful (and invasive) kind of location data entails tracking the change in location over time — called the path data. Mobile phones and sensors can track users/customers through a city or a store over the entire time they have their phones on or are in the store. A grocery store can track a customer through the aisles of the store, and such path data can help the store design its layout, determine where to stock items, or refine merchandising decisions.
Your iPhone tracks you when you leave your home, get in the car, and drive to work. By analyzing such path data over several days, machine-learning and AI techniques can learn your habits and provide you with personalized recommendations for the route to take to your office, alert you of traffic on the way, and suggest alternative routes.
The availability of path data can also help marketing firms personalize the experience for each customer much in the same way personalized display ads and offers appear to users on the internet. Merging online click-stream data with offline path data can help firms provide a seamless transition for customers as they shop online and offline.
Path data inside the home, collected and processed by internet of things devices, can provide enhanced amenities within living quarters — warming up the room as you emerge from the shower, decreasing the room temperature as you hit the bed, cooling the environment as you exercise on your stationary bike. The possibilities are endless. But this also means that users and customers give up some of their privacy in exchange for personalized service.
And therein lies the greatest challenge with location data for coming years: The way firms can deliver such services without compromising customers’ privacy.
P.K. Kannan is a professor of marketing science in the Robert H. Smith School of Business at the University of Maryland and a leading industry expert in marketing data, analytics and consumer behavior.