To know or not to know

Image: shutterstock

Some say it empowers people and democratizes data. Companies can’t wait to tap its seemingly endless potential. Still, there are quite a few naysayers who point to the potential threats of big data – not only to our privacy, but more importantly, to social relations and stability

By Beata Socha

After cloud computing, big data has been the buzz word of the business world for at least a couple of years. Big data means different things to different people. To some, it is the internet and social media; to others, it’s synonymous with digital data analytics. Whichever aspect of big data we discuss, we are bound to encounter both fervent enthusiasts and fierce opponents of the technology.
One thing is certain, it is a growing phenomenon and a fast expanding business.  The size of this year’s big data market is estimated at $16 billion. By the end of 2015, it will have expanded to $48 billion and 4.4 million jobs will have been created globally to support the industry, according to estimates by McKinsey Global Institute. Judging by these figures, it is clearly one of the fastest-growing industries, expanding six times faster than the overall IT market.

What’s not to like

Big data enthusiasts have quite a few arguments to support their claim that the internet, social media and digital analytics are improving the quality of life everywhere. For one thing, it is a powerful tool that can be used to organize people and ignite social change, the recent Arab Spring being one example.

“Today, social media and the internet are unquestionably accelerating change and empowering people who want to change their political system all across the globe,” said Amy Gershkoff, director of Consumer Analytics and Insights at eBay, as well as the architect of Barack Obama’s advertising strategy. According to Gershkoff, at the height of the protests in Egypt and Tunisia in March of 2011, a survey found that nine out of 10 Egyptians and Tunisians were using Facebook to organize protests or spread awareness about the protests. “Between January and April of 2011, the number of Facebook users in Egypt sprung 30 percent compared to the previous year,” she added.

Digital technologies are increasing political awareness also in developed, stable economies. “Nearly all Western democracies now have websites that allow citizens to see legislation under consideration by their government, including searching full texts of bills before they’re even voted on,” Gershkoff said.
As the head of the advertising strategy in Barack Obama’s both electoral campaigns, Gershkoff engineered a novel and extremely effective application of big data. “Using a neighbor-to-neighbor mobile application, Obama supporters could see which of their neighbors needed persuading to vote for Barack Obama and could knock on their doors and talk to them,” she explained.

Democratization of dataZrzut ekranu 2015-06-22 o 15.26.16

Another aspect of big data is digital analytics, the use of which “has democratized luxury, making it no longer the sole purview of the elite and wealthy,” as Gershkoff put it. It can provide services of a concierge, a personal shopper and a stylist to regular people.

Data science algorithms can review and analyze not only users’ behaviors based on their previous choices, but also the choices of people with similar tastes and preferences. In this way, mobile apps can recommend a restaurant, a trip or a movie that is tailored specifically to the needs of the user, just like a concierge service in a five-star hotel. The same goes for the selection of party music or an evening ensemble that fits the body shape and preferences of the person.

“Big data enables services that until just a few years ago were a luxury, to be delivered to regular people at affordable prices,” the eBay director added. She also believes that by providing more transparency to customers worldwide, big data will contribute to the proliferation of sustainable and green technologies in the future. A recent study by the Pure Research Center found that Millennials are the most sustainability-conscious generation yet. “Millennials are more likely to be willing to pay for products made by environmentally responsible firms that any previous generation, and the willingness to pay higher prices for green and sustainable products also extends to low-income Millennials,” Gershkoff said.

Do you know what you know?

Both global and local companies are already taking notice and spending sizable sums on learning more about their customers’ preferences. The demand for utilizing large data sets in decision-making is clearly growing. According to IBM, one out of every three business leaders doesn’t trust the information they use to make decisions. However, few realize that to get ahead in the game it is not really necessary to make major investments in data storage. According to Deloitte, 75 percent of senior executives from over 500 companies say that they are wasting more than half the data they already hold. To utilize the data
well they need competent people who not only know how to do it but also who can ask the right questions.

“The most important part of data analytics is not the data itself, nor the computers that run it, nor any of the technology – it’s the person doing the analysis. I’ve run teams of data scientists and statisticians in many jobs and the difference between good work and great work is the person who’s doing the work. It’s about combining creative thinking with scientific thinking,” said Gershkoff.
The job of a data scientist didn’t exists until just a few years ago. Today, there is an ever-growing demand for specialists who understand what large data sets can offer and how to use them to our advantage. McKinsey predicts a global shortage of 190,000 scientists capable of mining big data by 2019

A paradigm shift

What is it that makes these data scientists so valuable? What can they glean from the endless rows of ones and zeroes that is so important? Undeniably, analytics help companies make decisions more quickly, which is crucial in today’s social media-driven business. “Big data changes the paradigm that decisions should be made based on traditional statistics. Previously, a statistician made a hypothesis and tested it on a sample which verified whether the supposition was correct or not. With big data we are supposed to make decisions with absolute certainty,” said Arwid Mednis, legal counsel and partner at Wierzbowski Eversheds.

He went on to give an example of an American automotive company which collected a large amount of data from car manufacturers, dealers and mechanics. Based on the data collected, the firm discovered that the type of car least prone to accidents is a car painted orange. “In the era of big data, don’t ask why it is the way it is. The data tells you it is so and that’s that,” he concluded.
Another important aspect of big data is that the information collected never becomes outdated. Past data is and will continue to be used to make predictions about the future. Forecasts and projections are in fact one of the most crucial aspects of big data. “It is already used to predict where crimes will be committed. It can even be applied to predict what verdict the jury will reach,” Mednis explained.

Not all that glittersZrzut ekranu 2015-06-22 o 15.27.24

Collecting data about consumers and their preferences is nothing new. Retailers have been doing it for years. Based on their clients’ purchase history, they can easily predict what and when consumers are most likely to buy and send them a relevant enticement in the form of an ad or a coupon. Once big data algorithms compile the purchase history of all clients with similar preferences and tastes, the chances of a perfectly-tailored offer increases even further.
And that is exactly the bone of contention between big data enthusiasts and haters. One of the concerns most often raised about the use of big data is that the ominous “Big Brother” is already too “all-knowing” and fueling it with more information about us will only help confine people without their realization.
Mednis also gave an example of a teenager who was sent a pregnancy care package by a retailer a few years back because, based on the girl’s search entries, the company deduced that she must be pregnant. And indeed she was, but she wasn’t aware of it yet. To add insult to injury, the package was collected by the teenager’s father.

Illusion of anonymity

Anonymity is something we’ve all gotten used to thanks to the internet. It is supposed to keep the use of data troves within bounds. Opponents of he big data phenomenon claim that no one can be truly anonymous or forgotten once information is released. When Netflix released anonymous data of about 500,000 members of its club, it soon turned out that scientists were able to match the film reviews with actual members of the club. The controversy resulted in a class action lawsuit and a subsequent settlement.
This year, we witnessed Google lose its case against a Spanish lawyer, Mario Costeja, who demanded that his name be deleted from Google search results. He complained that entering his name in the search engine led to debt notices over a decade old. The European Court of Justice finally decided to favor the notion of privacy over the unencumbered flow of information, a decision which came as a big surprise to many.

Self-fulfilling prophecy

The risks that big data pose to privacy seem to go beyond receiving prompts and ads that seem to know us better than our spouse does. “Our banks already have detailed data about our payments and based on that data they can predict with 95 percent certainty which young couple will get divorced within the next five years,” said Maciej Ślusarek, attorney and partner at the LSW Leśnodorski, Ślusarek i Wspólnicy law firm.
The question is what happens when banks start using that knowledge when deciding about e.g. mortgage loan applications. “This could lead to a self-fulfilling prophecy. A bank that knows the couple will split up will deny them the loan. Then they will be forced to continue living in the cramped apartment they have, which will lead to a marital crisis.” The same quandary applies to insurance and medical data and how it can be used against regular people.
The problem with privacy is further exacerbated by the fact that we are not really sure how safe our data is. Even if the company does collect it and use it for the purposes we specifically consented to, there is still no guarantee that no one else will gain access to it illicitly. A recent study by consultancy EY revealed that most organizations (56 percent) admit that they would have difficulties with the identification and early detection of hacker threats. A third of them also said they don’t know how long it would take them to respond to these threats.

You say ‘tomato’

There seems to be one more ax to grind between the opponents and proponents of big data and its application – data profiling. We all know how profiling works. When two people enter the same phrase in a search engine they could get very different results, and different ads, depending on their previous queries. Enthusiasts will say: “What’s wrong with that? As a man I won’t mind not getting nail polish ads anymore.”
Meanwhile, opponents say that feeding people with information based solely on their previous choices will inevitably limit their understanding of the world. “Information profiling can lead to us stewing in our own sauce. We will only get information we want and the ads we want. Thus, no foreign stimuli will be able to reach us,” said Igor Ostrowski, president of the Council for Digitization (MAC) and Partner at law firm Dentons.
Ślusarek concurs. “We will read different things, compatible with our political views. Socialists will receive information only from left-wing sources and right-wing voters, from conservative sources. It will not bring us closer. It will likely make any understanding impossible,” he concluded.

As a matter of fact there is already a gaping abyss between how supporters of different political factions perceive facts, according to Radosław Markowski, head of the Democratic Studies Center at the University of Social Sciences and Humanities in Warsaw.

A study was conducted where supporters of the ruling Civic Platform party and opposition Law and Justice were confronted with plain and proven facts. When Civic Platform supporters were asked whether Poland’s crime rate fell in 2005-2007 (when Law and Justice was in power), most of them said “no.” Similarly, when Law and Justice supporters were asked whether Poland was the only country with positive GDP growth in 2009 in Europe (during the rule of Civic Platform), most of them also said it wasn’t true. “If we cannot agree on simple facts, how can there be any agreement on views and opinions, especially when we are fed polemic information over and over again?” Markowski mused.

Whether we think big data and its applications are a blessing or a curse, it is definitely here to stay. Once we tap the endless source of information, we will never be fully satiated. One question that data scientists and their algorithms probably cannot answer yet is: what will we do with all the data and where will it lead?

Pin It