North Carolina State’s Center for Innovation Management Research has been conducting research in Big Data analytics for the past nine years. With the help of many organizations—including the Clinton Health Access Initiative, Patheon, Kelly Services, Grifols, and Quintiles—CIMS has been providing a test bed to allow these organizations to collaborate on researching analytic applications aimed at improving strategic decision making.
CIMS Prof. Richard E. Kouri and data scientist Chad Morris are among those leading this effort. “Working together,” they report, “we have learned that the application of computer-assisted data analytics is rather straightforward—the real hurdles are people.” They elaborate in their article below.
The most common problem is the requirement for human interaction and the role that organizational culture plays in using analytics effectively. Senior managers, mid-level managers and technical personnel all need to share responsibility for developing a meaningful question, asking that question correctly, building an understanding of data-driven decision making, and creating a culture to implement these decisions quickly.
More than 1.5 million data-savvy managers are needed by 2018 in the U.S. alone, a McKinsey group pointed out (1). This is because the requisite skills and experiences have not become established practices within many traditional firms, and, importantly, are being taught in only a few universities (NC State being one).
Six Attributes of a Data-Savvy Manager
From our eight years working with more than 50 managers and engineers, we have identified the following attributes a good, data-savvy manager must possess.
- Be a good listener in order to understand the problem that senior or mid-level managers are facing.
- Be a good communicator who helps these managers place the problem into a set of questions that are actionable if answered; also, make sure the managers understand that decisions must be made once these questions are answered adequately.
- Understand the data analytics platform sufficiently so that actionable questions can be placed into a data model compatible with their platform.
- Understand how to evaluate the alternative approaches suggested by the data; be able to assemble the right team help prioritize them
- Help recognize and mitigate the inherent biases team members will have during the evaluation and prioritization process.
- Help managers with communicating the decision throughout the organization, putting the communication architecture in place and allocating the proper resources.
Working with Other Managers
The data-savvy manager will often find himself/herself allied with the other managers as they work to democratize the data-driven decision process throughout the organization. Employees, whether technically skilled or not, should recognize that data has the potential to impact every aspect of the business. The manager will be responsible for championing this message throughout the organization and begin shifting the organizational culture to one that is data decision-oriented.
A data-savvy manager must be able to structure a strategic problem and facilitate discussion in which the resolution or decision criteria are presented in an actionable fashion. As an example, this could be establishing the phase criteria for new product development to move to testing and validation or discontinuation of new product development. Using a project team of decision makers and subject matter experts, biases and alternatives to the decision criteria can be identified and formulated into a series of questions and sub-questions supported by the analytic platform and based on the decision criteria.
We have seen—and believe—that having the project team participate early in workshops for critical thinking and creative problem solving can help them offer insights into biases and identify needed resources upfront, thereby setting the projects up for success. Having the team involved should allow discussion to focus around the strategic problem and ultimately get its buy in on the plan moving forward. Gaining agreement early on is important for establishing accountability and responsibility throughout the process.
Strategic Questions Drive the Process
We stress that the strategic questions and sub-questions will drive the analytic process and must always be clearly tied back to the business need and desired outcome (2). Achieving early small wins can lead to big victories toward ultimately changing the organizational culture through the demonstration of value created for the firm and others.
In the planning phase, data-savvy managers must consider how to strategically advance the data decision process. By understanding the data platform and clear objectives for the data needs, the manager will be able to communicate effectively with the scientists or engineers and drive the needed data collection, modeling and analysis, based on the desired outcome.
Regarding the technical platform, we have seen a number of cases where the manager has made technological investments in their analytical stack based on the desired outcome and future planned applications. This was done to prevent costly investments in resources that might not provide a significant ROI and would add additional complexity to implementation and use. Having an understanding of the platform and the vision of what can be achieved, a data-savvy manager can identify opportunities to continue building out the desired analytical platform and the resources to further develop their data decision.
Maintain Data Collection
We have found that a critical component of the groundwork is to establish the practice of maintaining the data collection so that it can be applied to future work. The solution to the original questions almost always engenders additional questions. Thus, these data collections should become one of the regular tools that all employees can use when trying to address additional questions.
For example, the data-savvy manager uses these data collections, and some additional tools, in an iterative manner to orchestrate getting the right data to the right person at the right time. This is a tough, and also fun, job. Employing explicit data standards and processes early on can simplify the reuse of data for future questions.
Incorporate All Information
Lastly, it is critical that a data-savvy manager incorporate all of the information into the decision-making process. Whether it is information marketing will use for customer segmentation, or research and development for knowledge discovery, there must be an avenue for the information to be incorporated into the decision making process. A data-savvy manager must structure the process with the end in mind.
- “Big data: The next frontier for innovation, competition, and productivity”; McKinsey & Company; Dec. 5, 2015, http://www.mckinsey.com/insights/business_technology/big_data_the_next_frontier_for_innovation
- “Forget Data Scientists – Make Everyone Data Savvy”; Data Science Central; Oct. 17, 2015, http://www.datasciencecentral.com/profiles/blogs/forget-data-scientists-make-everyone-data-savv
- Richard E. Kouri, Chief Evangelist, CIMS, mailto:email@example.com
Chad Morris, Data Scientist, CIMS, mailto:firstname.lastname@example.org