Big Data: Big Hype, Utopian Promise, or Impending Hackocalypse?

Whether it’s big government or big corporations you fear may be encroaching on our right to privacy, you should have no trouble these days finding news to fuel your suspicions. If, on the other hand, you happen to run a business, you probably find clear-eyed assessments of the risks and benefits of things like cloud computing, Big Data, and the various emerging business intelligence technologies are really hard to come by.

Is the data you store in the cloud secure? The honest answer is that no data is ever perfectly secure—look what happened to Target. Any data connected to the internet is hackable, so unless you’re going to try to conduct business completely offline you just have to accept some degree of risk. Some tech experts even insist that the cloud is more secure than on-premises databases for many of the same reasons that money in the bank is more secure than cash in the mattress. Still, we’re just beginning to appreciate the possibilities of cloud computing, and the unfamiliarity is enough to make a lot of us uneasy.

At the same time, many businesses are rushing to take advantage of the cloud’s Big Data capabilities. If you think about Big Data’s potential to revolutionize industries like health care—patient information instantly channeled to every specialists in a hospital network, outcome statistics pointing the way to more effective treatments, solid data to inform public health initiatives—it’s easy to lose sight of the dangers to individual privacy. And this type of enthusiasm will likely draw many businesses into the Big Data game, since they know if they’re not taking advantage of the latest technology their competitors probably are.

The dirty little secret about Big Data is that we currently don’t know what impact it’s going to have on businesses or individuals. Though there are myriad ways to start easing Big Data into your business intelligence strategy, you also need to be careful not to rely too heavily on it. Here are three things to keep in mind:

Big Data isn’t necessarily good data (so don’t rush to conclusions).

Andrew Ross Sorkin cites the example of A&E’s decision to suspend Phil Robertson, one of the stars of the reality show “Duck Dynasty,” after he expressed his antigay views in a magazine interview. Right away, Twitter and other social media outlets registered the nation’s outrage at Robertson’s comments, and the network responded quickly to the trends in the data. The only problem was that the people who were going to Twitter with their grievances weren’t the same people who made up the show’s audience. After A&E had some time to analyze the source of the negative feedback, they ended up going back on their decision. Robertson never missed a day of filming.

Sorkin writes: “It seems as if just about every CEO of a global company these days is talking about how Big Data is going to transform their business. But with increasing frequency, it may be leading to flawed, panic-induced conclusions, often by ascribing too much value to a certain data point or by rushing to make a decision because the feedback is available so quickly.”

Having numbers isn’t the same as having insight.

The word moneyball, inspired by the book and subsequent movie of that name, is now used as a verb meaning roughly to statistically analyze performance to identify opportunities for improvement. The original story was about how the Oakland As used statistics to identify and recruit underappreciated players, a strategy that ended up paying huge dividends. Before long, every team was using metrics like this, and some were even creating new algorithms whose outputs are supposed to serve as key performance indicators, like UZR.

But, as Steve Mirsky points out, if Derek Jeter ranks in the bottom 10 percent in the league for UZR one year and in the top 10 percent the next, it raises doubts about what the metric is actually measuring. The bottom line is that, while numbers and statistics can be extremely useful, they also often create a false sense of certainty. Mirsky writes, “If all you do is count, you could tally up a million apples falling off apple trees without coming up with a theory of gravity.” In the business realm, you really have to keep track of where data is coming from, how it’s being structured, and what factors influence it if you want to make sound decisions. In other words, Big Data should be just one part of a larger business intelligence strategy.

Collecting certain kinds of data will make potential customers think you’re creepy.

Kate Crawford warns readers about what she calls “Big Data Stalking.” She cites the company Turnstyle as a chief offender in this category. Turnstyle puts up sensors that track smart phones as they search for a Wi-Fi signal and extract personalized information. When they aggregate the data to sell reports to businesses, helping them better “understand the customer” for more effective marketing, the identifying details are supposed to be lost. But there are two problems: first, the information is taken without the phone users’ knowledge or consent, and, second, people just have to take Turnstyle’s and similar companies’ word that they’re not keeping personal information or putting it to inappropriate uses. This is much the same problem many people have with NSA tracking programs; you can say that you’re only using “metadata,” but we know it wouldn’t be too difficult for you get more personal details.

The millennial generation is notoriously blaze about exposing personal details to the likes of Google and Facebook. But something tells me even the most trusting digital natives would be unhappy to discover the phones in their pockets had been hacked. What kind of policies we’ll need to put in place to protect privacy as more of this type of technology emerges remains to be decided. For now, though, it’s important to balance the prospective payoffs of procuring this kind of data against the potential hit your brand will take if the word gets out.

Big Data really does have a lot of potential to improve businesses and society in general. But you really have to understand what the numbers and graphs you’re looking at really mean if you’re going to make proper use of them. No matter how useful the metrics, you still have to consider them in the broader context of your organization and your customer base. Again, Big Data is definitely something you’ll want to check out, but it’s only one aspect of a comprehensive business intelligence strategy.

via Business 2 Community http://ift.tt/1hBXMAb