Types of Data Analytics

Data Analytics cannot be considered as one single solution to all the problems. There are different types in the analytics process, that are used to different problems in the business organization.

One of the important job as a data analyst is to find the right analytic type to be used in the right problem. Some of the analytic approaches have been used for a long time and some just recently. In the previous posts, we’ve seen in-depth data analytics and its uses. Now, let us see the types of data analytics, how and where it is used.

Descriptive Data Analytics

As the name implies, this type of analytics describes the data. It answers the question “what happened?”. It is a basic analytic technique that has been used in most of the organization for a long time. The main goal of this type is to analyze the past data to infer insights like “why the profits were low last year?” or “how satisfied the customer was of that product?”, etc.

It is used to learn about the overall performance of the business in the past. And it is a must for a business to study their past. With past knowledge, the business can make a better decision in the future. There are thousands of software to do descriptive analytics, and it doesn’t require an analyst to do it. A basic report or summary of the performance of a business is an example of descriptive analytics. It doesn’t use complex algorithms to compute. Basic aggregate functions like sum, average, count are enough to compute these analytics.

Google Analytics is a good example of descriptive analytics that gives insights into various aspects of your app or website.

Diagnostic Analytics

With descriptive analytics, you get to know “what happened?”. And with the diagnostic, you get the answer for “why it happened”? A successor to descriptive analytics, diagnostic analytics is used mostly always with descriptive analytics.

Based on the result of descriptive analytics, diagnostic analyzes further to find more insights to help the organization. It uses machine learning algorithms, data mining, and data drilling to find any anomalies or relationships in the data that could give further insights than descriptive analytics. The technique needs some extensive knowledge about the data and the algorithm. So, only an expert in the data field can work with the tools.

Predictive Analytics

A next step in the analytic process that tries to answer the question “what could happen based on the past?”. Organizations use this analytic technique to get insights about the future to make better decisions. This technique is still not widely used because of its complexity, but the software is growing to make this process an easy task.

Predictive analytics uses complex machine learning algorithms like linear regression, neural network, etc to compute the task of predicting the future. By analyzing the historical data, the algorithm finds the likelihood of different outcomes. The accuracy of the prediction is never 100%, but it varies depending upon the quality of the historical data.

All the major tech companies are now relying on this analytic technique. From Netflix to Amazon, predictive analytics is used to suggest possible products or movies to the users. Even financial companies are using this technique to predict the future price of the land, stock, or anything with value. Though it is not popular among small to medium businesses, that is likely to change in the coming decade because of the improvement in the software.

Prescriptive Analytics

The final and most useful analytic technique. Prescriptive Analytics works upon predictive analytics and gives further insights to the user on possible outcomes. It answers the question “what should you do?”. Yes, it gives knowledge for the business to take the best possible outcomes.

Predictive analytics just gives the probability of different outcomes, but prescriptive advice upon the outcome to take for the business to maximize profit or minimize loss. It uses different optimization techniques with the data and the business rules to simulate different possible futures to prescribe the best one.

As you can see, prescriptive analytics is a complex process and so, it is not yet popular even among some big organizations. Prescriptive analytics gives the most valuable insights if used. Once the business became aware of this technique and tools continue to grow, the prescriptive analytics will become popular.


As I said in the introduction, Data Analytics is not a single thing. It is comprised of various types. For a data analyst, it is important to know what analytic type to use when. Going into the next decade, the domain of data analytics is going to grow exponentially. Tools and Softwares are going to be developed that make the analysis task easier. Small to Medium companies will start to use the analytic process in their business, and data analytics will become even more popular than in the past decade.

1 thought on “Types of Data Analytics”

  1. Pingback: The Lifecycle of Data Analytics. » Pitch Engine

Leave a Comment

Your email address will not be published. Required fields are marked *