Financial Management and Business Data Analytics · Data Analysis and Modelling

Types of Data Analytics: Descriptive to Prescriptive

Updated 10 October 2026 · Fact-checked

Data analytics is commonly grouped into four types. Descriptive analytics shows what happened. Diagnostic analytics explains why it happened. Predictive analytics estimates what is likely to happen. Prescriptive analytics recommends what action to take. To answer a question, identify the question the business is asking and match it to the type.

Understand Types of Data Analytics

Data analytics means examining data to support a business decision. Different decisions need different questions answered. The four types are best understood as four questions a manager asks, in a natural order.

Descriptive analytics answers "What happened?" It summarises past data using totals, averages, percentages, tables and charts. A monthly sales report, a dashboard of branch-wise collections or a ratio summary of last year's accounts are descriptive. It is the most common type and the base for the others.

Diagnostic analytics answers "Why did it happen?" It digs into the descriptive result to find causes. Techniques include drill-down, comparing segments, finding correlations and variance analysis. For example, sales of a company fell in the western region. You break the fall down by product, month and customer and find that one large distributor stopped ordering.

Predictive analytics answers "What is likely to happen?" It uses historical data with statistical methods, such as regression, time series forecasting and classification models, to estimate future values or probabilities. Examples are forecasting next quarter's demand, estimating the chance a borrower will default, or predicting which customers may leave. It gives an estimate, not a certainty.

Prescriptive analytics answers "What should we do?" It goes beyond a forecast and recommends a best action, often using optimisation, simulation and rules. Examples are choosing the production mix that maximises contribution within machine-hour limits, setting a reorder plan, or deciding a delivery route that minimises cost.

The four types are often shown as a ladder of rising value and difficulty: descriptive, diagnostic, predictive, prescriptive. Real projects usually use more than one type together.

Key rules to remember

Descriptive analytics: Question: What happened? Input: past data. Output: summaries, reports, charts; Examples: sales dashboard, average cost per unit, branch-wise totals.
Diagnostic analytics: Question: Why did it happen? Input: past data split by factors. Output: causes and relationships; Examples: drill-down, variance analysis, correlation between two variables.
Predictive analytics: Question: What is likely to happen? Input: historical data and models. Output: forecasts and probabilities; Examples: demand forecast, credit default risk score, regression-based estimate.
Prescriptive analytics: Question: What should we do? Input: predictions plus constraints and objectives. Output: recommended action; Examples: optimal product mix, inventory plan, route optimisation.
Order of the four types: Descriptive → Diagnostic → Predictive → Prescriptive; Moves from hindsight to insight to foresight to action.

How to solve Types of Data Analytics questions

Most questions ask you to define the types, classify a situation, or suggest which type suits a business problem. Use the same method each time.

1Read the question and underline the business question being asked: what happened, why, what next, or what to do.
2Match it to the type: past summary is descriptive, cause is diagnostic, forecast is predictive, recommended action is prescriptive.
3State the type by name and define it in one line.
4Name the data and technique used, such as dashboards, drill-down, regression or optimisation.
5Link it to the business with a specific Indian example, using rupees or a named company type.
6If the question asks for a comparison, cover purpose, question answered, techniques, output and example for each type.
7Close with one line on how the types work together, for example descriptive data feeds predictive models.

Quickest way: Four-question shortcut

When to use it: Use for MCQs and for short classification questions where you have under two minutes.

Spot the key verb: reported or summarised means descriptive.
Why, reason or root cause means diagnostic.
Forecast, likely, probability or estimate means predictive.
Recommend, optimise, best course or should we means prescriptive.
If two options seem right, choose the one that matches the final output asked for in the question.

Common mistakes in Types of Data Analytics

Calling a forecast prescriptive because it helps decisions.
Every analytics type supports decisions, so students blur the line.
Fix: Ask whether the output is an estimate or a recommended action. Estimate is predictive. A recommended action is prescriptive.
Treating diagnostic analytics as the same as descriptive.
Both use past data and reports.
Fix: Descriptive reports the result. Diagnostic investigates the cause by drilling down or comparing factors.
Saying predictive analytics gives certain outcomes.
The word predict sounds exact.
Fix: Write that it gives estimates or probabilities based on past patterns, which can be wrong if conditions change.
Giving generic examples that fit no type clearly.
Students memorise definitions but not cases.
Fix: Keep one clear example ready for each type, such as monthly sales report, drill-down of a sales fall, demand forecast and product mix optimisation.
Listing the types in the wrong order or naming a fifth type.
Mixing with other lists, such as types of data.
Fix: Remember the sequence of questions: what, why, what next, what to do.

Worked examples

Example 1

A retail chain in Pune reports that its January sales were ₹48,00,000 against ₹55,00,000 in December. Classify each of the following as descriptive, diagnostic, predictive or prescriptive: (a) the monthly sales report above; (b) finding that the fall came mainly from one store whose freezer failed; (c) estimating February sales using the last 24 months of data; (d) deciding how many units of each product to stock in each store to maximise margin within warehouse space.

Show the solution

(a) The report summarises what happened in past months. It is descriptive.
(b) Finding the cause of the fall by looking into store-level data answers why. It is diagnostic.
(c) Estimating February sales from historical data answers what is likely to happen. It is predictive.
(d) Choosing the best stock quantities under a space limit recommends an action using optimisation. It is prescriptive.

Answer: (a) Descriptive; (b) Diagnostic; (c) Predictive; (d) Prescriptive.

Example 2

Explain how a private bank in India can use all four types of analytics in managing its personal loan portfolio.

Show the solution

Descriptive: the bank prepares reports on the number of loans, average ticket size and the percentage of overdue accounts by branch for the last quarter.
Diagnostic: it finds that overdue accounts are concentrated in one city and among borrowers in one employer segment, and traces this to layoffs at a few employers.
Predictive: it builds a model using past repayment data to estimate the probability that a new applicant will default. This is an estimate and not a certainty.
Prescriptive: it uses the risk scores and its limits on total exposure to recommend which applicants to approve and what interest rate or loan limit to offer.
Conclusion: the types work together, with descriptive data feeding the diagnostic and predictive stages, and the predictions feeding the final recommendation.

Answer: Descriptive reports the portfolio position, diagnostic explains the overdue pattern, predictive scores default risk, and prescriptive recommends approval and pricing decisions.

Exam tips

MCQs usually give a short business situation. Find the verb in it and match it to one of the four questions.
For descriptive questions, write the four types in order with one example each. This earns steps even if you forget details.
In comparison questions, use a small structure: question answered, techniques, output, example.
Always use a business example from finance or costing, such as sales variance, credit risk or product mix.
Do not mix this topic with types of data such as structured and unstructured. Those describe the data, not the analysis.

Practice questions from Data Analysis and Modelling

Types of Data Analytics in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Types of Data Analytics: frequently asked questions

What are the four types of data analytics?

They are descriptive, diagnostic, predictive and prescriptive analytics. They answer what happened, why it happened, what is likely to happen and what should be done.

What is the difference between predictive and prescriptive analytics?

Predictive analytics estimates future outcomes or probabilities. Prescriptive analytics uses such estimates along with objectives and constraints to recommend the best action.

Can you give an example of diagnostic analytics?

A company sees a fall in sales and drills down by region, product and customer. It finds that one distributor reduced orders. Finding that cause is diagnostic analytics.

Which type of analytics is used most in business?

Descriptive analytics is the most widely used because every organisation reports on past performance. The other types build on it and need more data, skill and tools.