CMU: Consumers Have Sharply <i>Reduced</i> Public Data Sharing

For years, conventional wisdom about privacy has been that shoppers—especially younger shoppers—have been consistently sharing more information online to the general public, a trend that would likely continue as privacy desensitization progressed. But a report released Tuesday (March 5) from Carnegie Mellon University found the opposite when it tracked 5,076 Facebook (NASDAQ:FB) users from 2005 through 2011, one of the most extensive studies of social media privacy yet.

"Over time, Facebook users in our dataset exhibited increasingly privacy-seeking behavior, progressively decreasing the amount of personal data shared publicly with unconnected profiles in the same network," the CMU report said. The implications for retailers are stark, suggesting that many of the privacy strategy underpinnings on both retail and e-tail may be flawed.

The report also found that those same consumers started sharing more information during that period, but only with people they assumed to be in a private group. And that sharing was expanded "both in terms of scope and amount of personal data." For retailers trying to extrapolate insights from this report to apply to chain CRM and mobile programs, these two conclusions are frustrating. Is public data comparable to posted comments on sites? And private data to answers shared directly with the chain? If this private-vs.- public privacy psychological dichotomy also plays out with shoppers, how would it impact shopper feelings about mobile tracking?

What does this all likely mean for retail? First, consumers are sharing more personal info overall—which is good—but they are decidedly not doing it because they have become desensitized to privacy in general. If that was the case, the study would have found an increase in public sharing and it, in fact, found the opposite.

The increases were with so-called private sharing. And with Facebook—and many other social sites—the definition of private is light years removed from what the dictionary considers private. Friends lists on Facebook can often move into the hundreds and many lists easily top 1,000. (Full disclosure: Please don't ask me how many of my LinkedIn connections are people I actually know. It's a touchy subject.)

And those increasingly large amounts of disclosed private data are not merely shared with that battalion-sized group of intimates. That data also is shared with, as the report notes, "third-party apps, advertisers and Facebook itself." And there's no practical limit on who Facebook will choose to share it with. Contractors? Partners? Random winos?

This suggests that shoppers put a great deal of faith in promises of privacy, even when those profound pretend privacy protocol promises are impressively not private. Hence, privacy policies need to be properly prominent.

This report truly reinforces what privacy officers have been arguing for years: Shoppers will share an awful lot, as long as at least a half-hearted effort is made to indicate that the data won't be shared with anyone, unless there's a critical business need to do so, such as if someone offers us a couple of bucks.

The most over-the-top experiment with social retailing was likely Blippy, a site that told everyone what a specific shopper was buying. That move was initially applauded by some retailers, until Blippys imploded in 2011.Making these distinctions even more challenging is the fact that the definition of what constitutes privacy is itself changing. At one time, for example, the question of whether a consumer is in a romantic relationship would have been the epitome of a private matter. For many Facebook fans who are teenagers or in their early 20s today, that is a very public declaration and deliberately so.

The report contrasted that relationship question with three other pieces of information Facebook has routinely sought: Instant Messenger name, phone number and physical home address.

"The first three elements (IM, phone, address) represent explicit vectors for communication, as in information that can be used to gain access to the person," the report said. "The fourth element (relationship) is a Facebook convention that signals a degree of willingness for communication. Looking for is shorthand for 'What type of a relationship are you looking for?' and responses can range from simple friendship to a serious relationship. Whereas the first three elements are vectors for contact information, 'Looking for' can be understood as a call-to-action for contact."

Perceived as a call to action, that data is something the consumer wants shared as widely as possible. Hence, what some might consider the ultimate in private data is shared enthusiastically while mundane datapoints such as a phone number are considered much more sensitive. And yet in what is believed to be a private circle, privacy is much more flexible.

The report also pointed out that, in the hands of an experienced cyberthief, all pieces of information can be pieced together to gather much more information. "The combination of hometown and birthdate information could be effectively used to predict an individual's social security number with reasonable accuracy," it said.

Privacy has always been this amorphous quality, which means different things to different people. And it means different things for those exact same people over time, as their income, age, marital status and professional settings change. That's why it's difficult to create a meaningful retail privacy policy.

When retailers who are pushing MCX talk about the reasons for such a retail payment alliance, one of their key points is privacy. But not the privacy of shopper data. It's about the privacy of CRM data going from one chain to another. It's ironic perfection that what they want to keep private are as many intimate details about their customers' shopping habits as possible. It's critical for them to have it, but an atrocity for a competing chain to have it. It's really not about privacy as much as having a competitive advantage.

Therein lies the real retail privacy conundrum. Most chains want to gather as much information as possible about their shoppers—surreptitiously, if possible—and to then ask those shoppers to volunteer much more data.

"Like a modern Sisyphus, some consumers strive to reach their chosen 'privacy spot,' their desired balance between revealing and protecting, only to be taken aback by the next privacy challenge," the report said.

That also nicely describes the retail privacy challenge. Shopper privacy promises have to be presented as comforting, a request that will never be used to help the chain take advantage of that information to persuade the shopper to buy more. When privacy policies are really legal phrasings of privacy violation policies, it's hard to make them sound convincing.

Thanks to CMU, there's good news. Those privacy declarations need not be especially convincing. They simply need to exist. Finally, a task that legal and marketing can easily master.