Strategic Cost Management · Introduction to Tools for Data Analytics

Data Analytics Basics and Four Types Explained

Updated 11 October 2026 · Fact-checked

Data analytics is the process of collecting, cleaning and examining data to find patterns that support decisions. It has four types: descriptive (what happened), diagnostic (why it happened), predictive (what is likely to happen) and prescriptive (what should we do). To answer questions, identify the question being asked, then match it to the type.

Understand Data Analytics Basics and Types

Data analytics means examining raw data to draw conclusions that help a business decide. Data on its own is just records: invoices, machine hours, purchase prices, sales by region. Analytics turns these records into information a manager can act on.

In cost and management accounting, the data comes from cost records, ERP systems, budgets, production logs and market sources. The scope is wide. You can analyse cost behaviour, spot cost overruns, forecast demand, set prices and decide product mix. Its importance lies in faster, evidence-based decisions, better cost control and early warning of problems.

The four types form a ladder. Each one answers a harder question than the one before.

Descriptive analytics: what happened? Example: total material cost last quarter was ₹48,00,000, up from ₹42,00,000. It uses summaries, averages, totals, dashboards and reports.
Diagnostic analytics: why did it happen? Example: the rise came mainly from a price increase on one input, not from higher usage. It uses variance analysis, drill-down and comparison.
Predictive analytics: what is likely to happen? Example: forecasting next quarter's demand or cost using trend and regression. It gives probabilities or estimates, not certainty.
Prescriptive analytics: what should we do? Example: recommending the product mix that maximises profit within machine-hour limits, using optimisation or simulation.

In a strategic cost management paper, the mark-earning skill is matching a business situation to the correct type and explaining how it helps a decision. Remember that the types are not rivals. A firm usually uses all four, and the later types depend on the earlier ones being done well. Poor data quality weakens every type.

Key rules to remember

Descriptive analytics: Question: What happened?; Summarises past data using totals, averages, trends and dashboards.
Diagnostic analytics: Question: Why did it happen?; Finds causes through drill-down, variance analysis and correlation.
Predictive analytics: Question: What is likely to happen?; Uses historical patterns, such as regression and time series, to estimate future outcomes.
Prescriptive analytics: Question: What should we do?; Recommends an action using optimisation, simulation and decision models.
Order of increasing value and complexity: Descriptive → Diagnostic → Predictive → Prescriptive; Useful for answering comparison questions.

How to solve Data Analytics Basics and Types questions

Most questions give a business situation and ask you to name the type, explain it or compare types. Use this method.

1Read the scenario and underline the verb or the question the manager is asking.
2Match it: reported or summarised means descriptive; cause or reason means diagnostic; forecast or likelihood means predictive; recommend or best action means prescriptive.
3Name the type clearly in the first line of your answer.
4Give a one-line definition of that type in your own words.
5Link it to the data and technique in the scenario, such as variance analysis, regression or linear programming.
6State the decision it supports, for example pricing, cost control or product mix.
7If a comparison is asked, cover each type on the same points: question answered, data used, technique, output.
8Close with a brief limitation or the next type that would follow.

Quickest way: Question-word shortcut

When to use it: For MCQs and short case-based questions where you must identify the type fast.

Look for the key word in the question.
What happened or how much: descriptive.
Why or what caused: diagnostic.
What will or likely: predictive.
What should or which is best: prescriptive.
If two seem to fit, choose the higher type only when the scenario asks for a recommendation or action.

Common mistakes in Data Analytics Basics and Types

Calling a variance report predictive analytics.
Students see numbers and assume a forecast.
Fix: A variance report explains past results. Finding why costs differed is diagnostic; just stating the variance is descriptive.
Treating predictive and prescriptive as the same.
Both look to the future.
Fix: Predictive estimates what may happen. Prescriptive recommends what to do about it. A demand forecast is predictive; the production plan chosen from it is prescriptive.
Saying predictive analytics gives certain results.
Forecast numbers look exact.
Fix: Write that it gives estimates or probabilities based on past patterns, subject to error.
Writing only definitions without a cost accounting link.
Students memorise textbook lines.
Fix: Add one cost example for every type, such as material price variance, cost forecast or product mix.
Confusing data analytics with data storage or reporting only.
The term is used loosely in practice.
Fix: State that analytics goes beyond storing or reporting data; it extracts insight to support decisions.

Worked examples

Example 1

A manufacturing company's monthly report shows that conveyor repair cost rose from ₹3,20,000 to ₹4,10,000. The cost accountant drills down and finds that most of the rise came from one ageing machine in the packing line. Identify the type of analytics and explain.

Show the solution

The report showing the rise from ₹3,20,000 to ₹4,10,000 is descriptive: it states what happened. The increase is ₹90,000.
The drill-down to find the ageing machine as the main cause answers why, so the analysis is diagnostic.
The main answer is diagnostic, because the question is about the cause.
Decision use: management can decide whether to repair, replace or change maintenance schedules.

Answer: Diagnostic analytics. The rise of ₹90,000 was first described, then the cause (an ageing machine) was found by drill-down.

Example 2

Distinguish between predictive and prescriptive analytics with one cost and management accounting example each.

Show the solution

Define predictive analytics: it uses past data to estimate what is likely to happen, using techniques like regression and time series.
Example: forecasting next quarter's sales volume and raw material cost from the last three years of data.
Define prescriptive analytics: it recommends the best action among alternatives, using optimisation or simulation.
Example: using linear programming to choose the product mix that maximises contribution within machine-hour limits, based on the forecast.
Compare: predictive answers what is likely; prescriptive answers what should be done. Predictive output is an estimate; prescriptive output is a recommendation.
Link: prescriptive analytics often uses predictive outputs as inputs.

Answer: Predictive analytics estimates future outcomes, for example a demand forecast. Prescriptive analytics recommends the action, for example the optimal product mix. The first informs; the second advises.

Exam tips

For MCQs, find the question word (what, why, likely, should) and match it to the type before reading the options.
In descriptive answers, always give a cost example; examiners reward application over definitions.
When asked to compare types, use the same four points for each: question, technique, output, use.
Do not present the four types as separate steps a firm must do in order; say they build on each other and are often used together.
Link analytics to decisions such as pricing, cost control and product mix to connect with the wider paper.

Practice questions from Introduction to Tools for Data Analytics

Data Analytics Basics and Types in other exams

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

Data Analytics Basics and Types: frequently asked questions

What are the four types of data analytics?

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

What is the difference between descriptive and predictive analytics?

Descriptive analytics summarises past data, such as last quarter's cost. Predictive analytics uses past patterns to estimate future outcomes, such as next quarter's cost. Predictive results are estimates and carry error.

How is data analytics used in cost and management accounting?

It supports cost control, variance investigation, forecasting, pricing and product mix decisions. Cost records, ERP data and budgets are typical inputs.

Is variance analysis descriptive or diagnostic?

Calculating a variance describes what happened. Investigating why it arose is diagnostic. In an exam, check whether the question asks for the amount or the cause.