Analytics and the Role of the CEO

In earlier posts I’ve expressed my perspective on how analytics projects should be run, and what factors make them successful and productive. Working with analytics requires a lot of technical skill to source, extract and integrate the right data. After that, the right statistical techniques need to be applied, and strong business domain knowledge is required pretty much throughout the lifecycle.

Apart from the technical skills, there are a lot of additional skills and attributes that a good team of data scientists should collectively have. Among them are the ability to communicate their results in visuals and language that can be understood by stakeholders outside of the team, particularly leaders who have to buy into the results. In addition to their core strength area (such as technology, statistics or business) it helps if every analytics team member also has at least some level of understanding and capability in the core strengths of the other team members. Team members must also be able to persevere patiently while working with Big Data, both while integrating it as well as while trying to actually analyse it.

It’s obvious that analytics involves executing a very focused set of tasks that require the team to keep their heads down and constantly look at all sorts of details. At the same time they still need to be able to zoom out to a larger view of the business in order to question or correlate what they see with events at a broader and more macro level. All their work is of course driven by a business question asked by some function or leader. But from how high up the organization does the drive need to come? I believe that analytics should ultimately be driven from the very top, ie, the CEO’s office. There’s a number of reasons for this.

The view from the top. A CEO that understands what analytics is all about would be best placed to ask questions in the right priority. While the analytics teams would have a great business understanding there is always the risk that at the level of detail that they work they might lose sight of the fullest picture of the business, or not consider some aspects of it. The view from the top is very different, it’s a 360 degree view of the business and it’s environment, and while the CEO does not get into all the detailed nitty gritty of analytics tasks they can certainly come up with the right questions to ask.

Sustaining patience. Analytics requires patience, and patience involves time and investment. In a running business and a dynamic external environment everyone always wants results in a hurry, but when results are not forthcoming it’s the CEO who can mandate sustenance of the investment and continued patience with the function.

Looking for disruptive change. Chances are that at the departmental or functional levels analytics will be used to achieve incremental change in the form of improvements to operating results. A CEO, however, may look beyond that, and constantly try to think in different ways about how the business can grow and perform better. This thinking can result in asking questions of analytics that can cause not just incremental change, but perhaps transformational change or even disruptive change.

Driving participation. Gaining new insights with analytics is one thing, but getting the business to participate whole heartedly and adopting analytics models in their operations may be hard. A CEO who believes that analytics could hold the key to improved business could make a significant difference in driving participation and adoption by asking for it to be applied across as many functional areas as possible.

Demanding a data driven culture. Although every business meticulously maintains books for accounts and uses various dashboard metrics to measure results, the majority of CEOs and other leaders admit to very often going by instinct to make decisions. But many CEOs do see the benefits in having their organizations become more data driven, and what better way to drive in a culture of being data driven than by asking that decisions and recommendations are based on analysis of the right data.