What your Facebook “Likes” Really Say About You
We all know that Facebook has a history of playing fast and loose with its users’ personal data. By now, anyone who uses the social network (in other words, almost everyone who surfs the ‘net at this point) knows that whatever data they put on the social site is likely to end up in the hands of marketers.
But, what if I were to tell you that it’s not just what you post or share, but what you do that could be revealing intimate details about who you are?
I’m not talking about visiting suspicious websites, or signing up for services or apps which are then given free reign of your personal information. I’m talking about the simple act of Liking a page, person, brand…whatever. What you Like says as much about you as what you actually write down and share. It all comes down to something known as unstructured or “Big” data, and the analysis of the information provided through it.
Analytics has become a pretty big deal in the business world of late. There’s a metric (measurement method, essentially) for pretty much everything that anyone’s ever thought might be important. Everybody’s convinced that data analysis will help business professionals make better, more informed, more intelligent decisions. As a result, we’re constantly on the lookout for new information with which to inform ourselves; more data that we can turn to our advantage. That’s where unstructured data comes in.
Whenever someone does anything online, be it accessing a website, liking a page, or purchasing a product, that action becomes part of a teeming sea of information. This collection of user activity can be referred to as unstructured data, and business professionals are convinced that if they sift through it long enough, they’ll eventually find something useful.
Moving back to the topic at hand, in this case, the ‘useful information’ comes from analyzing a particular individual’s likes. It’s all a matter of laying out as much information as possible on the table and poring over it for patterns and commonalities. Eventually, these patterns will emerge, and it will become clear which particular type of individual likes a particular product. The more things you publicly “Like” on Facebook, the more complete a picture a savvy marketing professional can paint of you.
With this seemingly innocuous information, they can make some frighteningly accurate predictions. Once the algorithm (or data scientist, if the marketers are so equipped) has noticed a correlation between two things – for example, most twentysomethings like to drink – it can use that correlation for the purposes of predictive marketing. Everything from age, to race, to income, to marital status can be predicted. It’s not unerring, but it’s accurate enough that these predictions are trustworthy.
The most distressing part about all this is that it’s not necessarily even a human doing all the work: it’s an automated system categorizing us based simply on what we’ve clicked the “Like” button on. I heard someone paraphrase a quote from Fight Club once, referring to Facebook narcissism. The quote went something like “you are not how many likes you have.”
Someone tell that to marketers. They have reason to believe otherwise.