With the prevalence of tools, software, hype, and so forth, one would think we whould have a clearer, not murkier, understanding of these topics. But, have we taken us further away from understanding? Why does it seem we are not gaining more traction now than in the past? This is the first in a series of blogs where we will tackle these topics. The overarching goal of data and analytics is to make smarter decisions and to help the business, so let’s get every organization to that point.
In the world of analytics, organizations for far too long have been throwing money at data and analytical software and tools, thinking it would be a magic potion, solving organizational problems and creating vast amounts of value. The problem is most organizations don't even have a grasp on the 4-levels of analytics, let alone a strong culture that allows for data literacy learning, data and analytical usage, and smart data-informed decisions. One step organizations can take to solve roadblocks is develop a sound understanding towards the 4-levels of analytics and how they interact together.
4-Levels of Analytics:
The first level of analytics is one everyone should be familiar with, and is one where organizations spend most of their time within the analytical sphere, but should really be just the first piece: Descriptive Analytics. To me, another phrase that can be used in place of descriptive analytics is observational analytics or reporting. This is truly the first level of analytics. Here is where organizations can utilize dashboards, reports, and so forth, to establish reporting and make observations from the data. This is also backwards looking analytics. But, this is truly the first level. What insight or "why" are we getting when we look at a data viz and see the line went up the last 4 quarters? None! All we are seeing is an observation. We need to take organizations to the second level of analytics, but unfortunately, they spend by far most of their time doing reports and dashboards...we have to help organizations progress past this.
The second level is Diagnostic Analytics. Another term I use for diagnostic analytics is insight. Organizations need to move past observational analytics, looking at reports, and empower the workforce to find the "why". Helping individuals understand that correlation doesn't mean causation, and empowerment to dig deeper than face value. This is where data literacy truly comes into play. When an individual is empowered with data literacy and an ability to use the third characteristic of its definition, analyze data, then individuals can dig in and find insight.
The third level is Predictive Analytics. If the first level is where organizations spend most of their time, this level is where they are spending most of their money. Organizations are hearing left and right they need AI, Big Data, and so forth. So, organizations are investing here. One issue with this thinking is the following: let's invest here, even though we spend most of our time in the first level and may not be very good at the second level, we are going to invest because we are told that is what we need. Now, organizations should be investing money in this level, but it is only a portion of investment. More should be invested in data literacy, so the whole organization can participate in the data and analytical strategy of the organization. When one thinks about it, organizations only have a few data scientists on staff, if that. Most investment should be to the empowerment of the whole organization with learning, data literacy, and the ability to get insights through other tools, such as a BI tool like Qlik Sense.
The final level of analytics we won't spend too much time on: Prescriptive Analytics. Prescriptive analytics is a fancy way of saying the data will let you know what to do. Now, this should be a target of organizations, to reach this level, but it should be part of an overall strategy.
Every organization across the world should be targeting and working towards embracing and utilizing the 4-levels of analytics. Organizations should stop spending the majority of their time in descriptive analytics, they need to succeed more in diagnostic analytics, should empower for proper use of predictive analytics, and finally, progress to prescriptive analytics. Instead of investing in magic solutions to solve the organization's data and analytical ills, organizations need to spend and invest in the empowerment of the workforce, building strong data and analytical strategies that tie back to the goals and objectives of the organization, and marry it the right tool investment and not just what sounds good.