Book Author and Speaker at TSAM Toronto
Is data scientist the sexiest job of the 21st century? Maybe for now.
In the October 2012 issue of the Harvard Business Review, Thomas Davenport and D. J. Patil called data scientist “the sexiest job of the 21st century.” Contrary to the popular opinion, data scientists are humans with a rare “combination of scientific background and computational and analytical skills” (Davenport and Patil, 2012). At the intersection of those skills are unicorns who score high in all three areas; they “are hard to find and are paid more than $200,000 per year,” argues Forbes contributor Gil Press (Press, 2015). The question is: how much more? We don’t know the answer for sure, because most unicorns have been captured, nurtured, deliciously fed, luxuriously lodged, and hidden from the public eye by dominant technology companies, such as Google, Amazon, Microsoft, Uber and Netflix.
The next important question is: what do data scientists exactly do that makes their job “sexy”? In his book Going Pro in Data Science. What It Takes to Succeed as a Professional Data Scientist, Jerry Overton argues:
I believe that data scientists are, foremost, scientists. They use the scientific method. They guess at hypotheses. They gather evidence. They draw conclusions. Like all other scientists, their job is to create and test hypotheses. Instead of specializing in a particular domain of the world, such as living organisms or volcanoes, data scientists specialize in the study of data. This means that, ultimately, data scientists must have a falsifiable hypothesis to do their job. (Overton 2016, 6)
Firms with limited surplus revenue will likely have a hard time buying into this value proposition. On the one hand, data scientists may want to work this way: guessing, gathering and drawing. However, it is unlikely that business people would want to have such unpredictable superheroes-partners at work.
Business is the organizing vehicle that gives meaning to the four creative disciplines: art, science, design, and engineering.
Figure 1. Business gives meaning to the four creative disciplines (adopted from Leurs, 2014)
Artists create powerful metaphors that inspire others’ imaginations. Scientists operate in nonpartitioned and unconstrained environment where everything is possible… in theory. Designers partition the problem space, define solution spaces, and introduce necessary constraints. Engineers make things real and precise. However, it is business that gives purpose to those activities by connecting creators with customers, organizing production, and delivering products and services to consumers.
Data science seems to be a complicated space with too many degrees of freedom and not enough constraints. And the most perceptive data scientists probably realize and relish this. A more predictable, and therefore more viable alternative to achieve business goals is advanced analytics. Leading technology research firm Gartner defines advanced analytics as a spectrum divided into four areas:
• Descriptive analytics provides historical hindsight into the company’s past performance and a view of the current operations in the form of reports, presentations, or dashboards.
• Diagnostic analytics, provides insight into causal relationships between events and outcomes and helps discover anomalies and emerging trends.
• Predictive analytics; provides foresight into possible future events and helps estimate achievable performance targets.
• Prescriptive analytics is responsible for determining the best course of action for a given set of situations.
Prescriptive Analytics: A Short Introduction to Counterintuitive Intelligence, the book that I have written with Noah Fang, describes an original framework for prescriptive analytics and provides practical guidelines to designing a prescriptive analytics platform and developing prescriptive analytics solutions (Milchman and Fang 2018). The framework creates synergy between analytics professionals and artificial intelligence algorithms by
• Clarifying and framing the problem space of prescriptive analytics;
• Defining a flexible and extensible solution space that takes advantage of plugin architecture and promotes reusability and composability of analytical components;
• Liberating analytics professionals from the complexity of machine learning algorithms and enabling them to focus on analytical work.
Prescriptive analytics aims to mass produce decision that can be used by autonomous artificial agents—robots, virtual agents and assistants, and smart Internet of Things objects—in trading and investing, enterprise fraud management, supply chain management, construction and manufacturing, and politics, where robots already prevail.
Now recall Marty Neumeier’s innovator’s mantra, “When everyone zigs, zag” (Neumier, 2006) and choose between zigging in the complicated, mysterious and adventurous world of data science and zagging in the more structured and predictable world of advanced analytics. No pressure…
P.S. I would like to thank Noah Fang, Michael Lapenna, and Jacqueline Flannery for their comments and suggestions.
Davenport Thomas H. and Patil D.J. 2012. “Data scientist: The sexiest job of the 21st century.” Harvard Business Review 90, 70–76.
Leurs, Bas. 2014. “Design Theory – Lecture 01: What is design?” SlideShare. Apr 17, 2014. https://www.slideshare.net/Leursism/design-theory-lecture01
Milchman, Andre, and Noah Fang. 2018. Prescriptive Analytics: A Short Introduction to Counterintuitive Intelligence. Toronto, ON: CreateSpace.
Neumier, Marty. 2006. The brand gap: how to bridge the distance between business strategy and design; a whiteboard overview. Indianapolis, IN: New Riders.
Overton, Jerry. Going Pro in Data Science. What It Takes to Succeed as a Professional Data Scientist. Sebastopol, CA: OReilly.
Press, Gil. 2015. “The Hunt for Unicorn Data Scientists Lifts Salaries for All Data Analytics Professionals” Forbes. October 9, 2015. https://www.forbes.com/sites/gilpress/2015/10/09/the-hunt-for-unicorn-data-scientists-liftssalaries-for-all-data-analytics-professionals/#2dd08a805258