- Key Chapter Points
- The Big Data Ecosystem
- The Defining Features of Big Data
- The Four Key Benefits of Big Data
- Lost in the Big Data Universe
- Big Data Without Borders
This chapter is from the book
This chapter is from the book
This chapter is from the book
The Four Key Benefits of Big Data
As we’ve said, Big Data and the digital economy is not a new topic; it’s been written about in the technical press since the phrase was first introduced in 1997. The three V’s—volume, velocity, and variety—widely used to describe the Big Data phenomenon were picked up and enhanced by Gartner as far back as 2001. But only recently has the Big Data discussion evoked the superlatives that we hear so often today. Consider, for example, what recent thought leaders have had to say about the subject.
The professional services group (PWC):
- As its potential becomes more evident, Big Data will transform every aspect of the organization, from strategy and business model design to marketing, product development, HR, operations and more....9
GE’s Joe Salvo, manager of the Complex Systems Engineering Laboratory at GE Global Research, stated:
- We are at an inflection point. The next wave of productivity will connect brilliant machines and people.10
McKinsey’s Global Institute estimates that:
- A retailer using big data to the full could increase its operating margin by more than 60 percent.... In the developed economies of Europe, government administrators could save more than €100 billion ($149 billion) in operational efficiency improvements alone by using big data, not including using big data to reduce fraud and errors and boost the collection of tax revenues. And users of services enabled by personal-location data could capture $600 billion in consumer surplus.11
Will Big Data really transform every aspect of the organization, deliver $600 billion in consumer surplus by capitalizing on personal-location data, and provide a typical retailer a 60% increase in operating margins?
Really?
If so, we’re looking at something important here. But the first step is obviously to understand what, exactly, we are talking about when we use the term Big Data, because it’s obvious that not everyone is talking about the same thing when they use the phrase.
The standard definitions and descriptions of the Big Data phenomenon help to understand what Big Data is all about, but don’t really help much when it comes to understanding why all this is so important, and why it is beginning to reshape our global economy now. For that, we need to look at the issue not in terms of what Big Data is, but rather, what Big Data does. And Big Data does four things:
First, Big Data Provides Unique Insight
As we’ve seen, Big Data is all about analyzing huge data sets to understand things in new ways—by using powerful computers to analyze a wide variety of source data to reveal hidden correlations and patterns in that data. Essentially, the mantra is “let the numbers speak for themselves.” There is a lot of evidence—from epidemiologists, economists, and even political pollsters like Nate Silver—that demonstrates that for those who are able to tap into the potential for analyzing large data sets, the insights can be profound.
Take, for example, the hospital looking after premature babies that now can capture data in real-time on every breath and heartbeat of each of the babies being cared for. All that data can be analyzed to predict infections 24 hours before the baby shows any visible symptoms.12 Or consider the Centre for Therapeutic Target Validation (CTTV) being created by GlaxoSmithKline and other bioscience centers, which shares early-stage research work that combines huge amounts of data on the biological processes behind disease and allows a variety of companies and researchers to analyze the data to look at how genetics can affect disease progression.13 Similarly, the National Weather Service uses Raytheon software to collect what will soon be as much as 5.4TB of data from US and other nations’ weather satellites each day—capturing data for every meter of the globe every four hours. It is a staggering amount of data that then needs to be combined with local information around the globe on temperatures, wind speeds, and barometric pressure, all analyzed using sophisticated algorithms and processed for everything from forecasts to assessments of sea ice concentrations.14
Unfortunately, getting this level of insight from huge amounts of data is not for the uninitiated. Numbers might not lie, but they can mislead. The complex algorithms behind large data set analysis are not easy to create and interpret, and statistical and data modelling issues can seriously distort conclusions. As we will see, although larger data sets do allow for more sophisticated analysis, that level of sophisticated analytics still remains largely in the fields of research and science, where data are much easier to control and manage, and skilled data scientists are in place to direct hypothesis generation and testing.
Whether that same level of insight can be easily achieved by the average manufacturer or retailer is yet to be proven. Many organizations should probably accept a much more prosaic reality, which is that most of the benefits that they are seeking—a better understanding of supply-chain costs and profitability or a better grasp of their customers’ buying patterns—could be achieved if they simply use their current technologies more effectively and apply more rigorous data-management techniques to the huge volumes of structured transaction data that they already collect.
Still, whatever its limitations, the unique level of insight that comes from complex calculations of large data sets is at the heart of the Big Data, and when academics, epidemiologists, statisticians, or economists characterize the Big Data phenomenon, this insight is almost always what they are talking about.
Second, Big Data Underpins Digital Advertising and Customized Individual Marketing
That same principle—that new technologies can be used to extract insight from large, unstructured data sets—can also apply to the fields of marketing and sales. In fact, when most retailers, advertisers, marketers, or business press columnists describe Big Data, they are most likely not talking about the advanced predictability calculations used by Nate Silver, GlaxoSmithKline, or the Federal Reserve. They are most likely talking about how companies can use large data set analytics and new storage and retrieval technologies to anticipate broader sales trends or as a means of capturing their customers’ personal data and creating customized marketing or sales messages. The principle is the same, and the methods and tools are similar, but the focus is different. These Big Data advocates are hoping that by gathering and analyzing huge amounts of information on individual customers they will be able to better target their advertising campaigns to sell more products.
Applying Big Data analytics to customer data has taken the retail world by storm. Consider a recent comment by luxury goods maker Burberry’s CEO, Angela Ahrendts: “Consumer data will be the biggest differentiator in the next two to three years. Whoever unlocks the reams of data and uses them strategically will win.”15
It’s an attractive idea. After all, the reasons why groups like Amazon or Google or Facebook have retained such high levels of funding and high stock valuations is that they have a huge advantage when it comes to applying their sophisticated search technologies to the data on millions of users of their systems. Although users may see these platforms in terms of the services they provide—a search engine, e-mail, a news feed, an online store, or a method for sharing photos—the company executives and their financial backers have always seen them primarily as platforms for collecting data and selling digital advertising. It took the world some time to realize that as remarkable as e-commerce and online search is, in reality, for the hugely successful Internet tech companies like Google, Facebook, or Twitter, it was always about the customer data.
As John Sargent, the chief executive of Macmillan, recently admitted in a New Yorker article about meeting Jeff Bezos at Amazon in the mid-1990s, “I thought he was just a bookstore, stupid me. Books were going to be the way to get the names and the data. Books were his customer-acquisition strategy.”16
If that strategy was unique to Jeff Bezos then, it certainly isn’t any longer. As we will see, the move toward capturing individual customer data and applying predictive analytics is all part of a fundamental change overtaking the advertising world, where the emphasis is shifting toward digital advertising and particularly toward digital advertising for mobile devices. And as more and more advertising revenues shift from print to digital, and from PC and TV to mobile, greater pressure is being placed on advertisers to more accurately target and deliver their ads. There is nothing more frustrating than a badly placed, poorly composed, or worst of all, an irrelevant, advertisement on a mobile device. The advertising world understands that if digital advertising is going to work—and the major media conglomerates and Internet platform leaders are betting their companies’ futures that it will—then those advertising messages are going to have to be much better made and targeted. And the best way to do that, they contend, is to understand all they can about individual customers.
There are other benefits. Obvious bottom-line efficiencies come from targeted ads, and particularly digital ads, compared with general, mass print mailings or e-mails. Marketers want to live in a world where the right message for the right product can be delivered to the right person at the right price at the right time—delivered and tracked for success on the customer’s mobile device. That, they say, has to mean increased sales and significant savings to an organization’s bottom line. And surely, they add, it must be beneficial for the customers themselves.
It all seems to be cause for celebration among the Internet tech companies, mobile app developers, and advertisers. Whether it is good news for brand owners or retailers is yet to be proven. Despite what we’re hearing from pro-Big Data marketers, it is difficult to determine yet whether this type of customized advertising really does add to a retailer’s top line (that is, more revenues through expanded sales). As we’ll see, although it may be more efficient to distribute targeted electronic marketing messages, at this point few studies show that this type of personalized advertising actually prompts customers to buy more, or differently, than before.
Still, advertising has never been an exact science. As John Wanamaker, once famously said, “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” Advertisers and marketers are hoping that by collecting extensive amounts of personal data on each of us in the future they will be able to do better than half.
Third, Big Data Creates a Market for Harvesting and Selling Customer Data
Even if a company doesn’t advertise and sell directly to customers through retail, it can still profit from Big Data by harvesting and selling its customer data. In fact, an enormous market for personal data has appeared that offers all types of organizations a tantalizing opportunity to sell what they know about their customers to others.
This capturing and selling client data can be eye-wateringly profitable. Walgreens, for example, admitted to the SEC that it sold its customers’ prescription and prescribing physician data to big pharma buyers in 2012 for $749 million.17 That’s the kind of money that makes executives sit up and take notice, and in the last few years, most organizations have considered the possibility of capturing and selling their own customers’ data. That was reflected well in a recent survey by Spencer Stuart of 171 US-based marketing executives, which found that marketers have widely embraced the idea of Big Data and customer data mining (see Figure 1.3).
)
Figure 1.3 Hoping Big Data Will Improve Marketing, Sales and Customer Service
Data Source: SpencerStuart18
Still, this type of customer data mining can go very wrong. A spate of lawsuits were filed against Walgreens (as well as CVS pharmacies) by customers angry about the retailer’s breach of personal medical data. Target used to boast of how it had been collecting and analyzing customer data for nearly a decade. But despite its efforts to leverage that data to improve customer experience, few Target customers would profess a sea change in their Target shopping experience during that time. Mostly, Target customers were not pleased to find that their personal data—including Social Security and credit card numbers—had been vacuumed up by hackers in a data breach that compromised 40 million payment card accounts. Target’s CEO resigned, the company spent more than $60 million in dealing with the breach, and its Christmas revenue went down 5%.19, 20
It is early days, and certainly too soon to predict where customers’ attitudes toward their data and privacy will go. Many contend that given the myriad and virtually unregulated number of sources collecting and selling our personal data, we might as well simply accede to the inevitable and admit that privacy as we once knew it is dead. But many believe that a customer revolt over data privacy issues is going to be the “blowback” of these customer Big Data policies, and as data breaches continue, we may see, through litigation or boycott or support for alternative privacy-ensuring software, customers create their own revolution when it comes to the use of their personal data. (Privacy issues are discussed further in Chapter 10, “Doing Business in a Big Data World.”)
Fourth, Big Data Supports Supply Chain and Industrial Services Efficiencies
A fourth important application of Big Data returns to the theme of insight but has nothing to do with customer profiling. Purists from the “make and move” industries believe that too much of the Big Data discussion is focused on collecting consumer data and developing customized advertising for mobile phones. They contend that instead, the world should be celebrating the ability of Big Data and new mechatronic components to create a revolutionary level of efficiencies in product development, production, or delivery. Industrial leaders from companies like Siemens or GE are interpreting Big Data as something very different from the marketers and advertisers, the research scientists or the economists. They see Big Data as new technologies to collect and analyze mostly machine-to-machine digital data throughout the supply chain, the Industrial Internet, and the coming Internet of Things.
The real benefit of Big Data, they say, comes from new machine-based self-monitoring and reporting technologies now appearing throughout the global supply chain. As these sophisticated sensors and interconnected diagnostic networks eventually merge with the Internet of Things, they contend that they will result in huge efficiencies (see Figure 1.4).
)
Figure 1.4 The Origins of Big Data Projects
Data Source: IDC’s 2012 Vertical IT and Communications Survey
Unquestionably, a lot is happening in the realm that GE dubbed the Industrial Internet, and advocates are right to say that early returns show that Big Data projects focused on reducing costs (the bottom line) through increased efficiencies have so far been a lot more effective than Big Data collection focused on expanding revenues (the top line) through customer profiling and customized advertising. In fact, bolstered by improvements in artificial intelligence and machine learning, the Industrial Internet might prove to be the most revolutionary in its outcome, because it will almost certainly have a profound impact in the coming years on productivity, employment, and the continued polarization of the economy.