What Are the Three Levels of Business Analytics? A Simple Classification Explained

Terms such as business analytics, data science, data mining, business intelligence and data analysis are now ubiquitous in both business and scientific circles. Their common goal is to transform data into actionable insights by leveraging rich data, traditional statistics and modern machine learning.

The rising popularity of business analytics can be attributed to four main drivers: demand for faster, better decisions; availability of data, powerful software and efficient algorithms; affordability thanks to analytics‑as‑a‑service models and cloud computing; and a cultural shift toward data‑driven decision making.

Simple Classification of Analytics

This article presents a straightforward three‑layer classification of analytics and describes each layer and their interrelationships.

Because of the strong demand for better, faster decisions and the increasing availability and affordability of hardware and software, analytics is growing faster than any other business function. Leading consulting firms and academic institutions are developing simple classification schemes to create a common context for analytics terminology.

Capgemini defines analytics as the reporting of data to identify trends, building models for prediction and process optimization, ultimately improving performance and achieving business goals. Executives view analytics as a core function that spans multiple departments; mature organizations embed analytics throughout the enterprise.

The classification divides analytics into descriptive, predictive and prescriptive categories. These layers are hierarchical but can overlap; organizations typically start with descriptive analytics, move to predictive, and finally adopt prescriptive analytics.

Three‑Layer Analytics Hierarchy

Descriptive Analytics is the entry‑level layer, focusing on reporting and business intelligence. It provides static snapshots of business activity and dynamic dashboards, often built on data warehouses designed for BI tools.

Predictive Analytics follows descriptive analytics and aims to answer “what will happen.” It includes techniques such as classification, regression, and time‑series forecasting to estimate future values of variables.

Prescriptive Analytics sits at the top of the hierarchy. It uses optimization, simulation, and heuristic models—many of which originated around World War II—to recommend the best actions, effectively answering “what should be done.”

The growing appeal of business analytics stems from its ability to provide decision makers with the information needed for success. The effectiveness of any analytics system depends on data quality and quantity, the accuracy, completeness and timeliness of data management systems, and the capabilities of the analytical tools and programs used.

This article is excerpted from Predictive Analytics: A Data‑Science Approach (2nd Edition) , published with permission (ISBN: 9787111718345). All rights reserved.