Companies need to tread a careful path through public sensitivity around the analysis and use of consumer data, warns James Dean
“Probably one of the coolest guys ever,” is Harper Reed’s description of himself. “I am an engineer who builds paradigm-shifting technology and leads others to do the same. I love using the enormity of the internet to bring people together.”
Mr Reed was Barack Obama’s chief technology officer in the run-up to the 2012 US presidential election. Data mining was his weapon of choice, drawing on the likes of Facebook, Twitter and various other online sources with Narwhal, his team’s purpose-built software platform.
With Narwhal, the Obama election team could find and mobilise voters that would have been unreachable a decade ago, and in the process raise billions of dollars of campaign money from millions of micro donations.
Hundreds of millions of snippets of data passed through Narwhal in the process. This is the “big data” that business is becoming so concerned with, although Mr Reed argues that the term big data is meaningless because the software of today is more than capable of sifting through vast volumes of information and rapidly providing a meaningful answer.
As the oft-repeated big data case study goes, Google can predict flu epidemics more quickly than doctors. It does this by finding patterns in what people are googling in certain areas, such as a spike in the number of people searching for “sore throat” in Kent.
Mr Reed and the Obama team were campaigning for an election, and Google’s ability to predict an epidemic is a useful public health service. Business, on the other hand, should not assume that public data is up for grabs for commercial gain.
“As is so often the case when new technologies and law collide, the latter is playing an erratic game of catch-up,” says Phil Smith, a director at law firm Pitmans. “The law on big data can politely be described as complex.
“By aggregating and analysing dozens of pieces of data gathered on an individual – each of which may in isolation be entirely innocuous – a product is created that is far more personal and sensitive than the sum of its parts; and, one might expect, deserving of greater legal protection.
“Privacy and data protection aside, there are numerous legal issues to consider, including intellectual property rights, competition law, contractual liability and regulatory issues.”
Notwithstanding the rules, the evidence suggests that consumers, on the whole, are averse to having their personal data trawled by companies
Politicians are also wary of the power of this new phenomenon. Edith Ramirez, the new boss of the US Federal Trade Commission (FTC), warned that her agency would be keeping a close watch. “The FTC will remain vigilant to ensure that while innovation pushes forward, consumer privacy is not engulfed,” she says.
Notwithstanding the rules, the evidence suggests that consumers, on the whole, are averse to having their personal data trawled by companies.
Target, a big American retailer, was able to accurately predict the delivery date of a pregnant customer and direct relevant adverts towards her. It made these predictions by drawing on various public data sources that, taken individually, were only vaguely connected to pregnancy. One of Target’s statisticians conceded in The New York Times: “If we send someone a catalogue and say, ‘Congratulations on your first child!’ and they’ve never told us they’re pregnant, that’s going to make some people uncomfortable. Even if you’re following the law, you can do things where people get queasy.”
A recent study by GfK, the research agency, suggested that only a third of consumers are happy to be targeted with adverts based on information trawled by companies. More than three quarters thought that it was “annoying” if companies got to know them too well.
“Not everybody is ill-disposed to companies using their personal data for targeting purposes, but they are in the minority,” says Colin Strong, managing director of technology at GfK. “Consumers often don’t feel as if they are part of an open dialogue about the nature of the trade that is being made. I think the advertising industry is in danger of thinking in terms of algorithms and programmatic opportunities while all the time losing sight of the fact that it’s a human being at the other end.”
But big data does not have to mean big advertising. “Big data provides an amazing opportunity to mine for new insights about human behaviour that brands can be using to their advantage,” Mr Strong says. “Social scientists are hugely excited about these opportunities.”
Fenwick, a French forklift truck manufacturer, installed sensors on its forklifts to gather information about how they were used. From this it developed a remote monitoring service and a school for forklift drivers.
And when BT trawled its customer prospects database recently, it analysed millions of records to determine who might like what services, enabling it to tailor its approach accordingly – and cut out unnecessary sales pitches.
Big data, therefore, should not be considered solely a means to target consumers with annoying, even creepy, adverts. The ability to identify patterns of behaviour and use these to improve business strategy has already been shown to be of great value.