Imagine having a doctor’s appointment, you’ll expect the doctor to ask for your symptoms, check your medical history, diagnose the ailment and proffer solutions before you consider that he’s done a good job, same analogy goes for Data Analytics. Data Analytics is quite broad, and it spans different aspects.

There are different types of analytical methods in Data Analytics which we’ll touch on shortly. While they are being itemized independently as Data Analytics types, they can also be seen as a logical progression of problem-solving with Data Analytics.

There are four (4) types namely Descriptive Analytics, Diagnostic Analytics, Predictive Analytics, and Prescriptive Analytics.

Let’s discuss them separately.

Descriptive Analytics As the name implies, Descriptive Analytics is a type of analytics that uses historical data, which is data collected over some time, to describe data. If all that’s needed to be known is an overview of your data without probing further, then descriptive analytics is what you need. It’s useful for research and reporting purposes. Excel and Tableau are two applications that can be used for descriptive analytics.

1. Diagnostic Analytics

After gathering information on what happened from descriptive analytics, the next logical step of action should be to know why it happened. Diagnostic analytics is used to probe further than descriptive analytics.

Checking data for trends and correlation are activities under diagnostic analytics and Data mining and regression are tools that will be needed for these activities. Tableau and Power BI can also be used to check trends using scatter plots and regression plots.

Under this type of analytics, we have bear in mind that correlation doesn’t mean causation. For example, we can say there’s a correlation between physical exercise and self-esteem; we however can’t state for sure that exercise causes self-esteem change without further investigation.

1. Predictive Analytics

Knowing what happened and why it happened isn’t enough, the next logical step is to predict the future. Using advanced mathematical techniques, a data analyst is meant to use the knowledge garnered from historical data to predict future trends with considerable accuracy.

For example, a company that has discovered that most of its staff that leave are the same people with a low satisfaction rate can predict that its employees with low satisfaction rates are most likely to leave also.

Machine learning, statistics, and modeling are useful for predictive analytics. You can also use Tableau to predict trends. This type of analytics is particularly useful for budgeting, risk management, and fraud detection.

1. Prescriptive Analytics

Imagine after a doctor’s appointment, the doctor only informs you about what’s wrong, why you feel how you feel, and feel you what will most likely happen if you continue on the same trajectory but doesn’t tell you what you need to do to either prevent or heal from the ailment. I bet you’ll feel disappointed and the same thought applies to a Data Analyst.

It’s not sufficient to be able to do the aforementioned types without being able to proffer solutions. This type of analytics proffers solutions to “what can we do to make … happen”. Prescriptive analytics will require hypothetical tests to be done; decision trees will also be useful to decide on the optimal course of action to be taken.

As a good Data Analyst, you have to be well versed in the various types of analytics as they build on themselves and have their roles to play in making a proper analytical decision.