Data Analytics isn’t a destination. It’s a journey.

One of the most important tools a company has in recent times and can have today, that can help them get insights into their business, is Data Analytics. These techniques are used to extract raw data, discover patterns, and derive valuable insights from it. Data analysis can help enterprises understand their customers, evaluate campaign performance, personalize content for respective audiences, devise content strategies, and develop new products in line with the trends. Businesses can leverage data analytics as a viable tool in enhancing business performance and improve bottom lines. 

In recent times, mankind is on a quest to digitize the world. This is leading to a huge surge in the global datasphere, a measure of how much new data is created and replicated each year. It is estimated that this will grow several times over the next few years. The total amount of new data created in 2025 is forecast to reach 175 zettabytes (ZB) from 33 zettabytes in 2018. [ref]

Such a large datasphere should not only be just collected or stored but also be enabled for businesses, enterprises, or individuals to make better decisions. This is the reason why many companies like Microsoft, Amazon, Google are heavily investing in such big data solutions. The big data industry, which started at a mere $7 Billion in 2011, was worth about $189 Billion by end of 2019, an increase of $20 Billion over 2018, and is set to continue its rapid growth and reach $274 Billion by 2022. [ref] 



What started as a simple conversion of data into summarized reports a few decades ago, with the further enhancements in the AI and ML technologies, Data Analytics has also undergone an exponential transformation. In the past, companies’ Data Analytics needs primarily focused on “What happened?” or “Why it happened?”. In recent times, the paradigm has moved to “What will happen?” and is moving towards “What would happen if the enterprise took this action?”. It’s evident that the trend has changed. From Descriptive and Diagnostic Analytics, we’ve moved to Predictive Analytics. Now, the trend is moving towards Prescriptive Analytics.  

Types of Data Analytics: How are they different?



  • Descriptive analytics: Data is used to describe what happened. For instance, the monthly sales reports or the annual expenses reports. This is the basic form of Data Analytics.
  • Diagnostic analytics: Data is used to describe why it happened. This usually includes drill down or across the data or mine the data to find the WHY. It is usually a successor to Descriptive analytics.
  • Predictive analytics: In this type, data is used to shed light on what happened and is used to describe what is going to happen. This uses advanced process models based on ML & AI. For instance, how my sales are going to perform in the next quarter, or how many digital bots would I be needing by the end of the year based on the current automation trends?
  • Prescriptive analytics: Data is used to provide information on not only just what will happen in the company, but also on how it could happen better if certain events have occurred. Beyond giving the necessary information, prescriptive analytics recommends actions that one should take to optimize a campaign, process, solution, or service to the highest level.

There is still a lot of confusion in these terminologies (Predictive & Prescriptive) and they are quite often used interchangeably. Nonetheless, each of these plays a very vital role in enterprises today.

How does Prescriptive Analytics help business?

Prescriptive Analytics makes it easier for businesses to make effective decisions. 

In the past, marketing teams would pull the historical data, analyzing and estimating who would be most open to receiving the product and build the campaigns around it. Many companies still do the same. This would be one of the most inefficient ways to go about. When one starts using Predictive analytics, things get easier. Machine Learning classification models can help us classify which groups of customers to target for promotions or which products to target for discounts, to maximize the impact. But the results of those campaigns are still unknown until the sales happen (or not).

For instance, a retail clothing store in the USA can see a spike in its sales during Black Friday every year. The same store when opening a new branch in Florida in November, can put a large collection of clothing and see a spike in sales. However, when analyzed using ML classification models and factoring the climate, a higher impact would be seen when focused on a few types of clothing rather than all. Now enters Prescriptive Analytics. Here Machine Learning models can guide the store by prescribing the right type of buyer for promotional offers at the right time, the right type of products to run offers on at the right price at the right time at the right location. This information allows us to maximize not just sales but profit overall.

The advantages of Predictive and Prescriptive analytics are not just limited to sales, but also to saving time, reducing transaction costs, and improving efficiency. These, if and when automated, can result in making real-time decisions. One of the best such examples would be the gasoline industry that changes the prices of their petroleum products throughout the day to maximize their profits. Achieving the benefits of predictive and prescriptive analytics comes down to having the right technology, infrastructure, systems, software, and processes. One should trust that the AI will do the work needed to maximize sales on one’s behalf, based on the historical analysis the algorithms are performing in the background, which is in turn driven by the systems of record, tools, and infrastructure. This requires entrusting control to the system. 

Although Prescriptive Analytics is quite powerful, this may not be needed by every company or campaign. Since the field itself is in its early stages, there is a long way for it to go with many tweaks and adjustments. It also takes a lot of time, effort, and data to make a powerful tool like Prescriptive Analytics, work. But having such a tool, in this competitive marketplace, would undoubtedly give a huge boost to the company’s productivity and profitability. The bottom line is, it is still quite early in the Prescriptive Analytics game and we are still scratching only its surface.


Which type of Data Analytics should you invest in?

To start answering this question, one should start with the most important one. “What do you want to accomplish?”. You can use the below guide on when to use which type of data analytics.


🧰 Tools Used Limitations ⏰ When to use

Descriptive & Diagnostic

What happened & Why?

  • Data aggregation
  • Data mining
  • Snapshot of the past
  • Limited ability to guide decisions
When you want to summarize results for business users


What might happen?

  • Statistical models
  • Simulation
  • Guess at the future
  • Helps inform low complexity decisions
When you want to make an educated guess at likely results


What should we do?

  • Optimization models
  • Heuristics
Most effective where you have more control over what is being modeled When you have important, complex or time-sensitive decisions to make

Once you know that, we can start with the Data Analytics journey and proceed through a series of phases that build on each other, transforming data from Descriptive & Diagnostic to Predictive to further Prescriptive.

At Vuram, our experts hold deep expertise in delivering Analytics services and can help your enterprise get started on its data analytics journey. Contact us today for a consultation; write to


– An article by Bhargav Parnandi, Architect, Analytics Division, Vuram




Submit a Comment

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

8 − 1 =

Pin It on Pinterest

Share This