Descriptive analysis in Microsoft Power BI – what happened?
Having data is half the success. Only the process of transforming them into information referred to as Business Intelligence can be used to increase profits, generate savings and increase market competitiveness.
The first question answered by data analysis is: what happened? Descriptive, i.e. description-based analytical methods detect regularities and quantitative relationships between historical data. The results of such a study can be used when making decisions about future events. A typical example of descriptive analysis are company reports that provide information such as the level of sales in a given quarter or the level of budget execution on an annual basis.
Diagnostic analysis in Power BI – why did something happen?
As the name suggests, diagnostic analysis is used to study the causes of a specific phenomenon in the organisation. Similarly to the descriptive method, it is based on historical data, from which it searches for those factors that influenced the occurrence of a given event. Thus, it provides information such as why sales in a given quarter were lower than planned or why the assumed annual budget was significantly exceeded.
Predictive analysis in Power BI – what can happen?
Predictive analysis, although like other methods, it refers to past data, tries to find answers to questions about future events. The data analysed in this respect provides information about possible scenarios and their effects. Such use of data is widely used – not only purely predictive but also practical. Particular benefits are brought by the ability to predict customer behaviour and preferences, risk management or verification of business hypotheses.
Prescriptive analysis in Power BI – what can be done?
Prescriptive analytics is also gaining in importance, whose task is to find optimal solutions to induce specific changes in business processes. In practice: the algorithms, taking into account factors known to them, predict the effects of particular actions at the prediction stage. At the next step, based on ready-made scenarios, prescriptive algorithms search for the best solutions. These can be used in the broadly understood optimisation processes, for example, concerning the sales channels of goods. And this taking into account many factors, such as production costs or storage capacities.
This type of analysis is perfect for all those places where, due to processing large amounts of data, it is necessary to implement automated decision making processes. According to Gartner analysts’ forecasts, included in the report “100 Data and Analytics Predictions Through 2020“, as many as 40% of companies will invest in the following years in prescriptive and predictive analysis.
