Source: Gartner
Many businesses today are aware of the great impact that data can have on the business’ sales and operations and are looking to implement analytics in their company.
In addition, with the increasing awareness to analytics, increased also the number of different analytics solutions and techniques, and many businesses try to figure out which solution fits them best.
In order to understand what is the solution that fits the business, I believe that the first step is to figure out what is the maturity level of the business in terms of analytics.
In general, there are 4 main categories of analytics:
- Descriptive analytics- answers the question “what happened”
- Diagnostic analytics- answers the question “why did it happen”
- Predictive analytics- answers the questions “what will happen”
- Prescriptive analytics- answers the question “what is the best that could happen”
Following is a summary of the different categories:
Descriptive Analytics
As the name implies, Descriptive Analytics describes the past- we analyse the past and try to understand from it insights that will affect the future.
The first step in descriptive analytics is data integration- integrate raw data from different sources, clean it, and create operational standard reports or Ad hoc reports.
As I see it, the descriptive analytics is actually the basis and foundation for the analytics capabilities in the business- the first stage should be understanding the past, and what happened, and getting insights out of the data.
Diagnostic Analytics
Diagnostic analytics is about understanding the reasons for the past outcomes- after understanding what happened, it helps us understand what are the reasons and causes for the results.
Diagnostic analytics uses techniques such as drill down, data discovery, data mining and visualization tools (such as dashboards). It identifies patterns, trends and segmentations, and applies statistical models and algorithms to understand correlations and connections between various variables.
Predictive Analytics
Predictive analytics is about predicting the future, it answers the questions of “what will happen” or “what is likely to happen”.
It uses techniques such as regression analysis, forecasting, multivariate statistics and predictive modelling, to analyse historical and current data and to forecast and predict future outcomes.
While there is no algorithm with 100% accuracy, the predictive analytics aims for the highest accuracy in the prediction models.
Some examples for the usage of predictive analytics in the business: predicting churn, LTV (Life time Value) of the users, fraud activity, sales and demand, etc.
Prescriptive Analytics
As the name implies, this type of analytics allows “prescribing” different actions- it answers the question of “what is the best that could happen” or “what should be done” by quantifying the outcome of future actions and advising on the best actions, based on the predictions.
Summarizing it in one word- prescriptive analytics is about Optimization– optimization of decisions and future outcomes.
The tools used for prescriptive analytics are business rules, algorithms, simulations, machine learning and computational modelling procedures. These tools use data from different sources, including real-time and historical data, big data and transactional data.
Between the different categories of analytics, the prescriptive analytics is the more complex one, and most companies are still not using these techniques.
However, this type of analytics can have a significant impact on the business.
To summarize:
- Descriptive analytics:
- For businesses that are interested in understanding what is going on in their business at an aggregate level
- For businesses interested in summarizing and describing some areas in their business (e.g. sales, customers, financials, inventory, etc)
- Diagnostic Analytics:
- For businesses interested in getting an understanding of the reasons to the past outcomes
- Predictive Analytics:
- For businesses interested in forecasting the future and predicting what will happen
- Prescriptive Analytics:
- For businesses interested in knowing “what should be done” and understanding future opportunities
As the business develops and improves its analytical capabilities, it becomes more proactive and less passive, and the value to the business also increases, as well as the competitive advantage.