Data Analytics: Getting Past the Easy Part (the Technology)

Everyone is seeking the holy grail of data analytics and, now, its derivatives – artificial intelligence and machine learning. The tools and platforms to accomplish this are powerful, and more programs across the country are offering training in data science and analytics skills.

Data AnalyticsSo what’s the holdup? Essentially, organizations are getting in their own way. In a recent analysis, Niko Mohr and Holger Hürtgen, both with McKinsey, observe that “the technical part is always the easier one to solve.” While data analytics is seen as the path to riches, “why is it that not everyone is already monetizing and achieving impact with it?” they ask. While the technical aspects are well developed and in place, most companies “are still significantly lagging behind when it comes to the structural parts such as getting the link or the translation between the technical and business worlds right, or managing the cultural shift away from gut-feeling-based decision making towards decision processes driven by data.”

Here’s what is at issue, the McKinsey authors state:

Data and business are in separate worlds

In too many organizations, data science and business teams are totally separate, Mohr and Hürtgen state. “This leads to a lack of understanding from the business side of what is possible and to the development of data science solutions that business doesn’t actually need.”

There’s too much of a gap between insight and impact

Many organizations do not take that final step of actually acting on the insights they receive. Companies may have bought into data analytics, and have even conducted one or more proofs of concept. “However, these proofs of concept are often isolated from each other and hardly ever turned into successful use cases, let alone scaled,” Mohr and Hürtgen point out.

Senior management doesn’t show enough commitment

Positive results only happen “when data analytics is implemented deep within and consistently throughout the organization. This requires the commitment and direction of a leader with the authority to drive this type of insights-oriented transformation, and many companies have yet to see that level of organizational commitment.”

Mohr and Hürtgen offer the following suggestions to get organizations out of their own way when it comes to data analytics:

Work “business backwards,” not “data forward”: First, identify business use cases you believe in and then think about the models and data you need to operationalize them, not vice versa,” they state.

Keep IT in the loop: “You might need to initially bypass corporate IT to start quickly and prove the concept early, but you will need to ultimately involve IT timely and heavily in order to scale,” Mohr and Hürtgen state.

Live and breathe agility: “A risk to fail small, to win big, a test-and-learn approach, and an experimental mindset should be woven into the organizational fabric.”

Empower users to act on analytics insights: Change management is crucial, the McKinsey authors observe. “Business users need to trust analytics results, and they need to be enabled to act on them.”