Beyond Big Data Skills: Creating the Right DNA for the Managers of the Data-driven Business World
Advice to schools and universities to help them prepare the future managers of the data-driven business world. One key step is to get managers acquainted with data by touching data, manipulating it, and 'playing' with it.
Kevin Carillo, Toulouse Business School.
The big data bubble has popped out revealing the inherent reality that lay within: the advent of a new business paradigm - data-driven business. Data is pervading organizations thus fundamentally changing their nature and functioning. As a consequence, the main mechanism that fuels the life of organizations, decision making, is being significantly impacted by the central role played by data. In other words, big data is gradually redefining the job of managers by incorporating data into decision processes.
For universities and educational institutions, the question is now to determine what they shall do in order to better prepare future managers to effectively evolve within this new data-driven business world.
A few years ago, when some organizations started realizing the potential of analyzing big data, they thought that a clever and talented individual could alone unleash the power of big data and do wonders. The quest for the mythical five-legged sheep that master statistics, computer science, and business, started; making the data scientist job one of the sexiest in the world. There is no doubt that there exists extremely talented individuals whose profile is close to the data scientist ideal. Nonetheless, examples showing that the ‘data scientist can do it all’ model does not work, abound. In this new data-driven business world, it is not one person that must be different but it is rather organizations as a whole and their overall functioning that must change.
New DNA and not new skills
There is a wealth of new technologies and tools around big data. Highest paying IT jobs ask for specific skills in NoSQL, Apache Hadoop, Python… Besides, starting from the function of Chief Data Officer, an array of new big data-related jobs have appeared in organizations such as data scientist, data analyst, data visualizer, big data solutions architect, big data engineer, big data consultant... to name but a few. However, by considering that behind the big data frenzy hides the looming shift towards a data-driven business era, the big data skill shortage goes way beyond the figures compiled by experts’ reports. In this new data-driven business era, big data must become a mindset that is ingrained into all employees’ DNA. This is all the more important for managers from any hierarchical level as no matter their decision scope (top, middle or operational), data and big data is becoming an inherent part of the inner mechanisms that govern the functioning of organizations: decision making.
Touching and playing with data to feel its business value
A sine qua non condition to put a big data mindset into our managers’ DNA is to get managers acquainted to data and make them apprehend the business value that is embedded into it. This can only be done by touching data, manipulating it, and ‘playing’ with it. Manipulating data is not a common task for managers who are more used to already digested and formatted information. Sensitizing future managers about the intrusion of data in the traditional decision lifecycle is key.
Whereas failure is used to be severely condemned in our nowadays highly competitive world, it becomes a positive mechanism in the big data world. Engendering an experiential culture within the mind of managers is a difficult task as it contradicts the basic nature of a manager’s job. Time is money and failing costs even more money. To convince (future) managers of the benefits of experimentation, the use of serious games or other types of simulation shall be recommended. Indeed, it is only through direct experimentation that one can seize that incremental learning allows to reach higher ends.
No way to run away from statistics and databases
Managers will not become data scientists. They will rather be an interface between a team of data experts (data scientists, data analysts, big data solution architect...) and higher management. They may also be directly involved in big data projects within their organization. As a result, it is necessary that all participating actors shall use a common language and have some shared understanding on a number of basic big data-related concepts. A necessary condition for schools and universities that are striving to diffuse big data into their curricula is to adjust the content of their business statistics courses and to emphasize the crucial importance of statistics in the big data context. Key database concepts (including managing and manipulating relational databases) shall also be taught as this provides a sound basis to understand important big data notions such as scalable distributed database systems or NoSQL.
Collaborating with the big data ecosystem
Big data is inherently complex. This results in an evolving and interconnected network of actors that interact (and often collaborate) with each other, covering a very wide spectrum of specialization domains such as applications (vertical, log data, ad/media), business intelligence, analytics, visualization, data/infrastructure as a service, analytics infrastructure, and even traditional structured database specialists.
From a pedagogical perspective, teaching big data is thus challenging. The complexity and many applications of big data can only be apprehended through direct involvement with the big data ecosystem. In simple terms, big data training programs, courses, and curricula shall be organized in such a way that students shall directly interact with an array of big data specialists in order to provide them a broad enough picture of the big data landscape.
Bio: Kevin Carillo is a learning enthusiast and big data advocate! He is an Associate Professor in Information Systems at Toulouse Business School.
- Free Advice For Building Your Data Science Career
- The Best Advice From Quora on ‘How to Learn Machine Learning’
- When Good Advice Goes Bad