Descriptive Analytics: Understanding the Past
Descriptive analytics is the foundational stage of the data analytics process, focusing on the examination and interpretation of historical data to answer the question, “What happened?” (Wolniak, 2023). It involves summarising past information to identify trends, patterns, and relationships that help organisations understand their performance and inform decision-making (Ojeda, Valera & Samadian, 2024).
Techniques include data aggregation, basic statistics, and visualisation, with outputs such as dashboards, reports, and KPIs (Rahman, 2025). These outputs provide stakeholders with clear and accessible insights, enabling them to assess progress, identify strengths and weaknesses, and monitor performance over time. For example, a sales report might reveal seasonal trends in product demand, while customer segmentation analysis could highlight the most profitable client groups (Wolniak, 2023).
In business reporting, descriptive analytics might be used to summarise last quarter’s sales or website visits. This information does not predict or prescribe but rather explains past performance.
Predictive Analytics: Looking into the Future
Predictive analytics builds on descriptive insights by using statistical models, algorithms, and machine learning to forecast likely future outcomes. By analysing historical and current data, it can identify trends, risks, and opportunities before they materialise (Eckerson, 2007). This enables organisations to make proactive, data-driven decisions such as anticipating product demand, identifying customers at risk of churn, or scheduling maintenance before equipment fails.
A defining capability of predictive analytics is its use of “what-if” modelling to test different scenarios and estimate potential impacts. For example, a retailer might analyse past purchasing patterns, demographic data, and customer engagement metrics to predict which shoppers are most likely to respond to a specific marketing campaign. These insights can inform targeted offers, improve customer retention, and optimise resource allocation.
Techniques such as regression analysis, decision trees, and neural networks are often employed to generate these forecasts. While descriptive analytics answers “what happened,” predictive analytics focuses on “what is likely to happen,” providing a forward-looking perspective that adds strategic value to data assets.
Prescriptive Analytics: Recommending Actions
Prescriptive analytics builds on the capabilities of predictive models by recommending specific actions that can improve future outcomes (Ncube & Ngulube, 2024). It uses techniques such as optimisation, simulation, and scenario analysis to evaluate potential strategies and select the most effective course of action under given constraints.
For example, in an educational context, prescriptive analytics might analyse student performance data, identify those at risk of underachievement, and recommend targeted interventions such as tutoring or tailored coursework. This not only forecasts possible results but also provides clear, data-driven guidance on how to achieve desired objectives.
The distinguishing feature of prescriptive analytics is its direct link to decision-making: while descriptive analytics explains what happened, and predictive analytics forecasts what might happen, prescriptive analytics answers the question, “What should we do next?”. By combining predictive outputs with decision models, it enables organisations to act confidently, balancing risks and potential rewards.
Action Point
Reflect on your current understanding and experience with data. Which of the three types of analytics do you apply most often in your day-to-day role? Explore opportunities to deepen your skills in predictive or prescriptive methods, as these are increasingly demanded in analytical roles.