Profiting from the Data Economy: Beyond Big Data
- Searching for the Next Generation of Quants
- From Big Data's Past to Its Future
- Characterizing Big Data
- Is Big Data a Strategy?
- Data Versus Insights
- Data and Value
- Value for Value
This chapter is from the book
This chapter is from the book
This chapter is from the book
While there has been a lot of discussion around the term “Big Data,” much of the discourse treats this as an abstract idea rather than a system in which we are all active participants. While the term has become ubiquitous, interest in the topic has not waned. A Google search for the term turns up approximately 13.7 million search results. A snapshot of Google Trends reveals the meteoric rise of queries for Big Data beginning in 2011 and increasing ever since then.1 Searching archived Twitter messages using Topsy reveals more than three million tweets referencing Big Data and in excess of two million tweets mentioning #bigdata.
Some contend that this marks a dramatic shift in what businesses and organizations are capable of doing. Others deride or critique it. Author and Duke University professor Dan Ariely likened Big Data to teenage sex: “Everyone talks about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so everyone claims they are doing it.”2 Whether you’re an ardent believer that the Big Data revolution is the best thing since sliced bread or you’re skeptical of the buzzword, there is no disputing the fact that more attention is being paid to the topic.
This attention isn’t just coming from corporate broom closets where statisticians are huddled over computers and poring over the output from complex analyses. Instead, data and analytics are garnering attention in the C-suite. In some companies, these topics fall under the purview of the CIO, whereas they are part of the CMO’s responsibilities for other companies. We’ve also seen the emergence of the chief data officer. Although you might expect to see this position at a financial institution or a company based in Silicon Valley, advertising juggernaut Ogilvy and Mather appointed its first global chief data officer in August 2013.3
The interest in capitalizing on the abundance of data extends beyond the boardroom to the public sphere. Microtargeting in political races was used as early as 2004. Local governments are also getting in on the act, with Philadelphia Mayor Michael Nutter naming the city’s first chief data officer in 2012 and New York City Mayor Michael Bloomberg appointing the city’s first chief analytics officer in 2013.
Searching for the Next Generation of Quants
Why are companies, campaigns, and governments focusing on individuals with a knack for data? Probably the same thinking that led Hal Varian, an emeritus economics professor from the University of California at Berkeley and chief economist for Google, to proclaim in 2009 that “the sexy job in the next ten years will be statisticians.”4 If you’re able to extract insights and act upon them, they can provide a strategic advantage. Across different types of organizations, statisticians can contribute immensely to improving operations, from increasing efficiency and cutting costs to increasing revenue. However, being well versed in statistics isn’t enough. What many organizations are seeking is a data scientist with the holy trinity of skills: someone with expertise in a particular field, coupled with knowledge of sophisticated statistical tools, and the technical expertise to develop and implement these algorithms on a large scale. This is often depicted as a Venn diagram (see Figure 1.1).
)
Figure 1.1 Data Science Venn diagram
Are they searching for mythical unicorns? Not necessarily. Both presidential campaigns in 2012 had chief data scientists. According to the vice president of Big Data products for IBM, a data scientist is “part analyst, part artist.”5 It’s not enough to crunch numbers in the background. The findings from advanced analytics are only as good as the way in which the insights are communicated to key decision makers. When we talk about using marketing analytics to guide strategy, we’re not just talking about a set of curve-fitting exercises. Instead, we’re talking about storytelling, informed by data, which has the potential to inform decision makers.
Are these three skills all essential, or can we get by with someone who is lacking one of them? With knowledge of statistics and the ability to code, a researcher can make data sing. However, although this may be sufficient from the standpoint of conducting research, what practical decisions can they support? Without a sufficiently deep understanding of the domain in which they’re operating, the impact of the insights on strategy will be limited. Meanwhile, someone who understands statistical models and knows the domain in which they operate is valuable from the standpoint of producing insights, but they are limited in their ability to convert those insights into a scalable solution. We similarly run into problems if individuals are fluent computer scientists but lack knowledge of the statistical models that are often used for evaluating business decisions.
Universities are making an effort to address the significant talent gap. Many have started to offer programs that tap into the interest in Big Data and data science. North Carolina State University, through its collaboration with SAS, launched the Institute for Advanced Analytics and offers an M.S. in analytics. At New York University, you’ll find the Center for Data Science offering an M.S. in data science, as well as an M.S. in business analytics offered by NYU Stern School of Business. The marketing department at the Wharton School of the University of Pennsylvania houses the Wharton Customer Analytics Initiative, while the operations and information management department offers a track in business analytics. Northwestern University offers an M.S. in predictive analytics through the School of Continuing Studies, while Northwestern Engineering houses the M.S. in analytics. You’ll also find an M.S. in analytics offered by the University of San Francisco and an M.S. in marketing analytics at the University of Maryland.
From this small sampling of the programs and the initiatives that have developed in higher education around analytics, you can start to see why it’s difficult to prepare students for the roles that organizations are seeking to fill. There’s a fundamental question about where the appropriate training for dealing with data takes place. The two logical schools in which programs focusing on Big Data would emerge are business schools, where the insights have the potential to guide decisions, and computer science and engineering departments, where the technical tools may take center stage rather than strategic decisions.