AACSB Blog

student arms/hands on laptop with colorful data icon overlay

An Insatiable Demand for Analytics Talent Is Reshaping Business Education


Posted February 07, 2018 by Dan LeClair - Executive Vice President and Chief Strategy and Innovation Officer - AACSB International

Five years ago Thomas Davenport and D.J. Patel wrote in the Harvard Business Review that data scientists “are akin to Wall Street ‘quants’ of the 1980s and 1990s.” Experts in physics and math streamed into investment banks and hedge funds, where they designed new algorithms and data strategies. This initial wave was followed by new university-based financial engineering programs that “churned out a second generation of talent.” The authors predicted the same would happen in business analytics.

They were right. The number of university-based analytics programs has since exploded. Although it is difficult to pin down a precise figure due to the absence of universally-accepted definitions and terminology, AACSB’s database of programs now lists more than 400 analytics degree programs across nearly 220 business schools worldwide.

A broader view, however, reveals that a lot more is happening. In addition to universities, companies such as Microsoft and IBM offer education and training in data analytics. These programs are accessible online, often through open providers such as edX, and can lead to certifications. Many large companies are also involved in large-scale efforts to reskill or upskill their own employees to think and act differently using data analytics.

Then there is Kaggle, a platform supporting competitions in analytics that Google’s chief economist once described as “a way to organize the brainpower of the world’s most talented data scientists and make it accessible to organizations of every size.” With its global community of “Kagglers,” active online forums, and access to employers (through recruitment competitions), it is hard to escape seeing Kaggle as an education provider.

So there is more than meets the eye in the unfolding story of business analytics education. This article traces some of these developments in order to provide insights for talent leaders, as well as business educators, about where analytics education is heading—and how it is changing business education. The article is also intended to serve as a high-level primer to support AACSB’s upcoming Data Analytics Summit, which will consider strategies for achieving AACSB’s objective of accelerating the development of business analytics education.

AACSB Data Analytics Summit March 19-20, 2018

Connecting Academia and Practice

Business analytics involves a distinctive combination of knowledge and skills. Any data analytics practitioner is expected to have certain technical skills. In addition to a having a strong foundation in math and statistics, they should be able to use specific tools—such as Jupyter Notebook, Apache Hadoop, and R Studio—and apply those tools in practice. According to practitioners, however, technical skills are considerably more valuable when combined with knowledge about how business works. Such knowledge helps data scientists empathize with users, who are other managers across their organizations, and bring intuition to the task of identifying patterns and developing new, profitable insights.

Analytics professionals also need general management skills. As one data scientist wrote, “I thought that getting a job as a data scientist was all about proving that you’re the smartest, most technical, most experienced person in the room. ... What it’s really about is showing that you can work collaboratively with the rest of the team, work fast to keep up with business needs, pick the right tool for the jobs at hand, and that when you build things, you can answer a question that impacts the bottom line of the company.”

What strikes me about this list of requisite knowledge and skills is that it points directly to the value connecting academia and practice. Academic providers, such as business schools, want access to businesses in order to stay abreast of the tools and techniques being used in practice, to gain access to raw data, and to provide exposure to business situations in which analytics skills are applied. Meanwhile, company-based programs built on practice can be enriched by tapping into the experience of business schools in providing broad-based business knowledge and developing core management skills.

One way that business schools are incorporating practical training is by providing learners with access to certification training to complement their degree education. While degrees are trusted and portable, they are notoriously opaque when it comes to signaling specific technical competencies. Certifications, on the other hand, are viewed as more transparent in representing technical competencies achieved.

Business schools are also starting to bundle learning modules into smaller packages, which will enable new and different combinations. New systems, such as Degreed, are facilitating comparisons and connections across credentials, certificates, specializations, badges, and the like. As a consequence, we expect to see new partnerships between business and business schools, such as the duel-certificate program offered by Microsoft and the Kelley School of Business at Indiana University. I also see huge potential for platforms, such as Kaggle, to connect with other educational programs, helping to crowdsource data for course projects as well as provide a supportive learning community.

Breaking Down Silos

Silos can be problematic in any organization. A common problem in analytics initiatives, data silos are repositories of fixed data controlled by a single department and isolated from the rest of the organization’s data. In large companies the legacy data systems of subsidiary enterprises are a common cause of data silos. Continuing the theme of skills versus knowledge, coding skills are needed to integrate the data, while business knowledge and experience are needed to identify the silos.

Higher education has its own silos. They exist in the institution between schools and faculties, such as engineering and medicine, as well as inside business schools between scholars in different disciplines, such as information systems and finance. These silos have developed over decades and are embedded in the organizational structure. In business analytics practice, however, the problems are not confined to the boundaries of university and business school silos. So it is no surprise that disciplinary silos have hindered the ability of universities to respond quickly to the demand for analytics talent.

Scholarly success in academe puts an extraordinary premium on specialization, motivating scholars toward narrower subjects and raising the barriers to integration. The rewards in business schools favor publishing scholarly articles in academic journals. And the structures and traditions that reinforce these rewards are powerful. I once heard a PhD data scientist at a leading internet company confide to an audience of executive education leaders that his former academic colleagues didn’t care at all about business—they only cared about publishing articles.

As I have written elsewhere, I’m nonetheless optimistic that focusing on technology, and analytics in particular, can help universities to break disciplinary silos, or at least make them less relevant. Developing analytics programs have challenged different faculties to work closer together. Business schools are motivated to partner with other schools, integrate curricula, team-teach courses, and guide student projects with professors in other areas. And scholars are starting to moving more readily between academia and practice, where they work on problems that cross disciplines.

To use one example to illustrate the potential for change, the Insight Data Science Fellows Program trains PhDs from a wide variety of disciplines, including physics, mathematics, and engineering, with the tools and perspectives to be effective data scientists in industry. In 2014, the program was receiving more than 500 applicants for each class of 30 students, and had a 100 percent placement rate into data careers afterward. It will be interesting to consider how these scholars move between practice and academia, as well as how they can potentially serve as business faculty in the future.

Linking to Ethics and Purpose

Going back to the comparison of data scientists to Wall Street quants, there are increasing concerns about the underlying ethical and moral dilemmas associated with data, especially as it advances into artificial intelligence and automation. After all, the quants have been blamed for the financial crisis of 2007–08 (though unfairly in my opinion). The ethical and philosophical questions embedded in the changing analytics landscape can be quite complex and important, affecting privacy and human rights, for example.

I worry about these issues because it seems to me that analytics education has been driven mostly by two opportunities: monetizing data and fostering employability. Both matter, of course, but as we develop analytics talent, we cannot ignore the need for learners to identify and address ethical issues. For this, a broader education is needed.

When I visit business schools, I always encourage them to think beyond employability when developing any program. Employability of graduates is necessary, but thinking about the overarching purpose gives the program a center of gravity or context that connects with the school’s mission. In business analytics, it might as simple as wanting to develop “data-driven leaders,” or it might be connected to a discipline, such as “democratizing data for marketing,” or to an industry, such as “powering health analytics.”

There are signs that ethics and purpose are becoming more important in analytics education and training, in business and academe. Apple University has an ethics professor and philosopher on staff. The University of Chicago’s Data Science Social Good program trains “data scientists to tackle problems that really matter.” And there is an initiative to develop a code of ethics for data scientists.

I see great promise in the variety of educational programs for developing analytics skills, and not only to meet rapidly expanding industry demand. Guided by the distinctive combination of competencies and skills and the cross-disciplinary nature of the problems, we are driving academia and practice closer together and breaking silos in businesses and universities. By letting larger society questions guide our development, we are leveraging analytics for social good—and destined to succeed in transforming business education for global prosperity.


Dan LeClairFollow Dan LeClair on Twitter @AACSBDan.

Share:

‚Äč