SAN JOSE, CALIF. — Starting off the second day of the 2009 Where 2.0 conference, Alex "Sandy" Pentland, a professor at the Massachusetts Institute of Technology, opened with a presentation that was indicative of those to follow: how location-based technology can improve the way people exist in the physical world. His talk, "Reality Mining for Companies: How Social Networks Network Best," examined how location and proximity to individuals enables us to create a literal "map" of social networking. In doing so, organizations can study precisely how individuals are interacting with each other and how that correlates with productivity and creativity.
According to Pentland, people have characteristic ways of acting depending on whether they're in a work, home, or play environment. He's found that observing these patterns on the social graph will generate predictions regarding who will share with whom that are twice as accurate as predictions developed simply by asking those individuals in a survey. For example, analysts can observe social influence by determining whether the presence of Person A in one location indicates the likelihood of Person B being there as well. "When you see me, it's good odds the other guy will show up as well," he said. This notion of statistical connection between individuals relies on what Pentland called "conditional dependence data," which helps unite segments of social relationships. The application, he says, goes beyond just social networking; rather, the patterns point to the power of "directed social networking," which predicates social dominance and influence.
The applications of this capability, though still in the development stage, suggest that analysts can predict buying behaviors, eating behaviors, and even political views of interconnected consumers. This information, Pentland says, "is the core of a lot of businesses and the heart of customer relationship management."
Pentland conducted an experiment that examined the electronic (email) and physical (face-to-face) interactions of employees at a German bank that was about to launch an advertising campaign. The location-based social graph monitored the activity of managers, developers, and personnel in the sales, support, and customer service divisions. Just by looking at the flow of digital and analog communication, Pentland identified immediately that no one was talking to customer service, a department that is understandably critical to the development of a consumer-facing initiative.
Another study followed employees in their day-to-day interactions. The results indicated that while increasing an employee's access to information by one standard deviation can boost productivity by 4 percent, increasing her face-to-face time with other employees improved productivity by 10 percent, and as much as 40 percent in creative productivity. Face-to-face interaction, he concluded, is an important source of social support. Moreover, daily interaction enables the exchange of "tacit knowledge" within an organization (e.g., the "rules" of the corporation, or employee work behaviors).
For the many companies that are undergoing "restructuring," Pentland said, no single method has emerged for determining whether such reorganizations generate the intended effect. On the other hand, he noted, with a simple dashboard, executives would be able to identify which employees are truly communicating, and which ones remain stuck within a traditional silo.
At the community level, Pentland cited a similar study that took place across the United Kingdom. Communities were monitored for their social patterns and the same general principle applied: Communities that displayed a high rate of interactivity were more prosperous (i.e., lower crime rates, higher rate of employment) than those that weren't. Social patterns can also be used to identify suspected criminals. Pentland, who lived in India for several years, reported that the Indian government had determined that 80 percent of the terrorist-related crimes that were resolved in the last two decades involved this type of location-based social patterning.
"The fundamental [objective] is moving from [a] static view of the world to a dynamic view of how an organization really works," Pentland said. Demographic information (e.g., age, gender, income) has provided limited insight in the effort to target individuals. The examination of dynamic information -- where people are going, what they're doing -- can reveal that two 35-year-old men who live in the same neighborhood, work in the same building, and even shop in the same grocery store are actually very different. One may respond to diaper ads, while the other may react to news of a new bar opening.
As a final thought, Pentland emphasized that privacy remains a pressing concern. Consumers, he suggested, should hold the complete rights to their information, even when they voluntarily provide the data. Given that location technologies are currently in their early stages, particularly from a B2C perspective, regulations have yet to be explicitly defined.
"You should all be creeped out," Pentland said. "If you're not, you weren't listening."
News relevant to the customer relationship management industry is posted several times a day on destinationCRM.com, in addition to the news section Insight that appears every month in the pages of CRM magazine. You may leave a public comment regarding this article by clicking on "Comments" at the top; to contact the editors, please email editor@destinationCRM.com.