We should treat data in an organization as our asset. It is sometimes said that nowadays data for companies, is the modern equivalent of oil, that is, it represents a great value in business. They are the ones that directly tell us what's going on in the business, what it currently looks like, what state it's in, where it's going, and how we can use it most effectively. It goes without saying that good organizational data management and analysis requires specific skills and tools. It is the agility of an organization in using data and its positive effect on its development that we call analytical maturity of an organization, which is having an increasing impact not only on technology companies, but also on other industries that are becoming increasingly digitized.
There are quite a few definitions of analytical maturity, and in many management publications you can find another one treating selected aspects of organizational maturity. One of the simple yet popular divisions that speak of the analytical maturity of an organization is the ability to use successive levels of analytics:
- descriptive,
- predictive,
- prescriptive.
The prerequisite, let's call it the zero level, for talking about analytics in an organization, is to have consistent and reliable data, available to a set of people analyzing it. This seems rather obvious, nevertheless, problems with "finding" data in organizations are not that uncommon again, not to mention having a publicly available platform for processing it, which, by the way, is already often considered a higher level of analytical maturity. Nevertheless, it's important to remember that it all starts with data availability.
Once you get the basic (zero) analytical level, i.e. "we have the data," you need to know it thoroughly. A fairly natural first step is to simply describe the data, or simply see what's in it, more formally called descriptive analytics, which is actually the first level of analytical maturity for an organization in this view. Even with a small amount of data, in fact, manual checking is of little use, so we harness mathematics and statistics, analyzing the data through simple measures like averages, medians, standard deviations and so on. At this stage we also often start building reports and visualizations, as numerical measures alone may not be telling enough. This will allow us to actually understand what is happening in our historical data and how our business is behaving.
The natural next level of analytical sophistication will be to try to predict the future from the data, or predictive analytics. In this analytics, we use mathematics and statistics to create numerous, potentially quite complex models that, based on current and historical data, allow us to predict the behavior of an element of the business in the future. Based on the results of the predictive models, we have additional information about how our business will behave in the (near) future, which makes it easier for us and helps us make decisions that affect the future of our business.
Before I move on to the third level, let me point out that business representatives are very good at predicting and typifying trends, but they have trouble explaining how they came to their conclusions, i.e. quantifying their experience. This is where prescriptive analytics enters the scene, allowing us to see why this is happening in our business. It will also help us learn what needs to change if we want to influence our customers even better. This is the highest level of maturity that any data-driven organization should strive for. It is worth mentioning here that the distinction between predictive and prescriptive analytics is sometimes quite arbitrary, and it is often debated which category a given technique belongs to. Nevertheless, in the approach described here, a company in category three that knows how to use prescriptive analytics in its business uses data to analyze its past, with its various variables, to make changes and strategies in the organization aimed at its most optimal functioning in the future. In addition, it knows how to monitor changes in the business environment on an ongoing basis and respond relatively nimbly to them.
More about analytical maturity in an organization is discussed in classes at the postgraduate programs on Data Science and Big Data in Management. They prepare students to manage an organization on the basis of technologies related to Data Science and Big Data. Graduates will be able to effectively carry out the transformation of enterprises or departments into a data analytics-based enterprise. Recruitment for the fall edition of these studies is currently underway.
Interesting links:
