© 2019 SourceMedia. All rights reserved.

The Hierarchy of Needs for Analytics

Analytics is a buzzword that is transforming boardroom discussions into scientific data-backed debates. That’s mostly a good thing – the problem is, most people don’t really know what they are talking about. We see too many organizations that achieve success with one minor analytics project and then try to live off that glamour, like middle-aged suburbanites reminiscing about that epic night out in college.

What these companies really need to do is scale their analytics efforts – turn that one success into the first of many. In order to do that, there are series of steps an organization must take, and certain needs that must be met. You may be familiar with Maslow’s Hierarchy of Needs, developed to explain the needs of the human race in pyramid form, from the most basic to the most advanced. In that spirit, we’ve developed our Hierarchy of Analytics Needs. Study this model to see what’s required for the Analytics equivalent of Self-Actualization (the ultimate in the analytics world? Artificial Intelligence, of course).

Level 1 – Questions/Curiosity

In Maslow’s original, physiological needs like food and water are the most essential requirements for human survival. In the analytics world, questions and curiosity are food and water. Keeping the system fed with the right questions and the right framing for problems are vital. A health analytics group will ask more questions than it answers.  

To create and foster this in your organization, ensure that a specific group, if not the entire organization, has the mandate to only ask - and ask more questions about everything. Encourage the habit of refining questions and relooking at questionable decisions. Never settle for easy answers and do not let your analytics group become an answering machine.

Level 2 – Data

With the physiological needs relatively satisfied, in Maslow’s world, an individual’s safety needs take precedence and dominate behavior. Safety in the analytical world comes in the guise of data, which actually enables decision making. When Sherlock cried, “Data! Data! I can’t make bricks without clay,” he was indeed pointing out to one of the most primitive needs for any analytical undertaking – data. Vulnerability at this level is maximum for the young and uninitiated and that’s why a lot of organizations tend to spend a lot of time and effort in procuring data even before engaging in decision making. Don’t let unavailability of data or poor data quality become a roadblock here. Appoint a Data Governance team that is mandated with two things – smartly translating any available data into good usable data, and unifying all the data sources and data ingress points into one common data platform. The power of a unified, cohesive and usable data platform cannot be underestimated.

Walmart is the master of data. When they don’t have it in-house already, they find a way to get it. When its analytics team didn’t have the essential quality of transactional data they needed to predict stockouts, they used historical weather patterns at the zip-code level. The team also uses Pinterest and Facebook data for merchandising decisions, to assess which products and trends are hot. Think outside the box to acquire or create the data needed to solve the problem at hand.

Level 3 – People

After physiological and safety needs are met, humans need love, belonging and other interpersonal connections, said Maslow. Similarly any analytical enterprise needs the right amount of love and belonging from people who have an ecosystem to provide that analytical love.

A classic mistake that organizations commit at this level is the fallacy of ‘expertise’. While the Infinite Monkey theorem could be an absurdity on one end, so would be the attitude of hiring experts, on the other end.  In the fast changing world of analytics, where one key technology today might become irrelevant in the next two years or where a totally new technology needs to be mastered in a short span of time, the right kind of people would be those willing to learn, master and unlearn those technologies at a sustainable pace. And don’t forget the most important asset: an insane amount of curiosity.

Level 4 – Socialization

In Maslow’s model, Esteem presents the typical human desire to be accepted and valued by others, and that drives humans to engage in activities that gain recognition. Quite similarly, an analytical enterprise in an organization needs the recognition to survive and thrive beyond being an elite group of number crunchers. This can happen only via widespread socialization and evangelization of analytical outputs and their associated insights. Remember that consumption of analytics is more important than creation: any organization that creates more reports, analyses and insights than that are consumed suffers the inevitable death-by-numbers.

Successful socialization can happen through a top-down outcome, behavior and insight approach to problem solving. Be sure that the outputs and insights from your analysis can be understood by senior leadership as well as the rank and file. One organization that does a great job at this is NYC Analytics, a team within the New York City government. Their analysis of Hurricane Sandy through a Disaster Response and Resiliency template not only helped allocated response resources for the right folks at the right time, but helped city planners understand the ramifications of the hurricane.

Level 5 – Artificial Intelligence

“What a man can be; he must be!” said Maslow, and this sentiment applies to any analytical enterprise as well.  Different organizations might have differing needs from an analytical enterprise, but the basis for that self-actualization would arise out of the ability to relegate the quotidian/mundane tasks to automation, and to use people to identify the patterns that disrupt conventional business thinking. To understand this level of need, the analytical set up must not only achieve the previous needs, but also master them efficiently. Artificial Intelligence is eventually the answer to the immortality of an analytical set-up.

Some of the best examples of organizations that have achieved analytical nirvana would be Google and Amazon. Google has developed some of the world’s smartest machine-learning algorithms that can predict to the exact word what a human being is going to think next. Similarly, Amazon’s all-powerful recommendation engines seem to know a person’s tastes better than their own mothers.

Can every organization achieve the penultimate stage of Artificial Intelligence? No. And not every person can achieve Self Actualization. But as with Maslow’s original theory, the higher you get to the top of that pyramid, the more content you probably are.  So the journey itself still has a lot of value.

Kshira Saagar is a client engagement manager at Mu Sigma.

For reprint and licensing requests for this article, click here.