Stages of Analytics
Analytics is commonly broken down into four progressive stages. Each stage answers a deeper question than the one before it, and together they form a natural maturity path for how organisations use data.
1. Descriptive Analytics - "What happened?"
Summarises historical data using reports, dashboards, and basic statistics. Example: "Total sales last quarter were 12% higher than the quarter before."
2. Diagnostic Analytics - "Why did it happen?"
Digs deeper into the descriptive findings to identify causes, often using techniques like drill-downs and correlation analysis. Example: "Sales rose because of a festive-season discount campaign."
3. Predictive Analytics - "What is likely to happen?"
Uses statistical models and machine learning on historical data to forecast future outcomes. Example: "Based on past trends, sales are expected to rise by 8% next quarter."
4. Prescriptive Analytics - "What should we do about it?"
Goes a step further to recommend specific actions, often using optimisation techniques. Example: "Increase inventory for the top 3 selling products by 15% ahead of the festive season."
Coming Up Next
Next, you'll look at a widely used framework for running a Data Science project end to end - the CRISP-DM life cycle.