From Descriptive to Prescriptive: Exploring the Pillars of Data Analytics

Today, companies have access to many types of data because of advances in technology. This data can include sales numbers, customer feedback, or supply chain details. Companies study this data to find useful insights that help them make informed business decisions.

This is where Data Analytics comes in. It means looking at data to find helpful information. The term "meaningful insights" is very important here. These are insights that answer key business questions. To do this, different tools and methods are used.

Data analysts use different methods to understand data and help businesses make informed business decisions. These methods also help them find useful information that companies can act on.

There are three main types of data analytics models:

• Descriptive

• Predictive

• Prescriptive

Each model works in a different way and has its own purpose. Let’s explore each of these models to understand their differences and how each of these can be used to enhance business decision-making process.

Descriptive Analytics

It is a process of summarizing historical data to reveal past performances, patterns and trends.

It answers the question, ‘What happened’.

Tools and Techniques used:

Ø  Data Aggregation – used to gather and collate data from various sources.  One example of the tool used in this technique is excel.

Ø  Data Mining – Determining patterns and relationships from large datasets.

Ø  Data Visualization – Utilizing Business intelligence platforms like Power BI / Tableau to generate dashboards and scorecards.

Ø  Statistical Analysis - Applying statistical methods to summarize and describe data.

Example:

Dashboards showing trends in GDM incidence.

Healthcare providers review dashboards showing trends in GDM incidence by trimester, patient age groups, and BMI categories.  These insights help guide nutritional counseling programs, screening schedules, and resource planning for endocrinology referrals in the next quarter, thereby improving maternal and fetal outcomes.

Predictive Analytics

This involves the use of historical data and statistical algorithms to forecast future events. Primary goal of this analytics is to predict future outcomes based on past trends and patterns.

Predictive analytics answers the question, "What could happen next?".

Tools and Techniques used:

Ø  Regression Analysis – to understand the relationship between dependent and independent variables.

Ø  Time Series Analysis – Analyzing data points collected or recorded at specific time intervals.

Ø  Machine Learning – means teaching computers to learn from data and make predictions. It uses special methods called algorithms, like decision trees, and works with tools from Python programming libraries to

do this.

Example:

 Predictive analytics to find out which pregnant patients might be at risk of gestational diabetes (GDM).

A maternity care center uses predictive analytics to find out which pregnant patients might be at risk of gestational diabetes (GDM). They look at past patient data like age, BMI, family history of diabetes, ethnicity, and early pregnancy blood sugar levels.

Doctors get alerts on their screens when a patient is at high risk. This helps them act early in their pregnancy instead of waiting for problems to happen. It leads to better care for both the mother and baby and helps use resources more efficiently.

Prescriptive Analytics

Prescriptive analytics suggests what to do next to get the best results. It uses data, algorithms and business rules to recommend the best actions to take. It answers the question, "What should we do?"

Tools and Techniques used:

Ø  Optimization: Finding the best solution from a set of feasible options.

Ø  Decision Analysis – Assessing and comparing different decision options.

Ø  Machine Learning – Using algorithms to learn from data and make recommendations.

Example:

Prescriptive analytics to give personalized advice to pregnant women who might get gestational diabetes (GDM)

A maternity care center uses prescriptive analytics to give personalized advice to pregnant women who might get gestational diabetes (GDM). After predictive models flag patients likely to develop GDM, the system suggests what to do next, like joining a special diet program, meeting with a diabetes doctor early, or checking blood sugar at home. The system also helps plan appointments and use resources wisely, so doctors can focus on the patients who need the most help.

Prescriptive analytics doesn’t just predict—it tells healthcare teams what steps to take. It turns data into smart decisions, helping them act quickly and confidently.

Conclusion

Descriptive, Predictive, and Prescriptive analytics are key types of data analysis. Descriptive analytics helps summarize past data, Predictive analytics uses that data to forecast future outcomes, and Prescriptive analytics recommends the best actions to take.

Together, they help businesses make smarter and more informed decisions.