Types of Analytics (Descriptive, Diagnostic, Predictive, Prescriptive)
Part of the Data Analyst Roadmap
In short
Not every data question is the same kind. "What happened?" and "What should we do?" need very different work, and mixing them up wastes days.
Descriptive answers 'what happened', diagnostic answers 'why it happened', predictive answers 'what will happen', and prescriptive answers 'what should we do' — most analyst work lives in the first two.
Think of it like this
A cricket match, four questions
The scoreboard says India made 280 (what happened). The replay shows the middle order collapsed to spin (why). The win predictor says 65% chance (what will happen). The captain decides to bowl spin early (what to do). Four questions, four kinds of analytics.
Watch it happen
Four questions, rising difficulty
1/5Descriptive: what happened? "Sales were ₹42 lakh in March." Sums, counts, dashboards.
The big ideas
What you need to know
Descriptive is the daily bread
Weekly reports and dashboards are descriptive. Boring sounding, but every other type depends on them being right.
Diagnostic needs digging
To explain why, you slice data by region, product or time until the change shows up in one place.
Predictive guesses the future
It uses past patterns to forecast. Always share it as a range with uncertainty, never a promise.
Prescriptive recommends
It compares possible actions and their likely results. A human still makes the final call.
Save yourself the pain
Common mistakes
Oops
Calling a monthly sales chart 'predictive analytics' to sound impressive.
Do this instead
Name it honestly. A chart of past sales is descriptive, and that's perfectly useful.
Oops
Jumping to predictions before you understand why the past numbers moved.
Do this instead
Do the diagnostic work first. A forecast built on a misunderstood trend will mislead.
Your mission
Sort the questions
Quick check
"Which customers are likely to stop ordering next month?" is which type?
Go deeper
- 1Classify 4 real reports or dashboards you've seen into these categories.
- 2Notice how much value comes from just doing descriptive and diagnostic well.
- 3Learn why prescriptive analytics usually needs much more mature data infrastructure.
More topics at this stop
What Data Analysts Actually Do
Data Analyst vs Data Scientist vs Data Engineer
The Data Analytics Lifecycle
Data-Driven Decision Making
Common Industries & Domains for Analysts
A Day in the Life of a Data Analyst
The Data Analyst Career Ladder
Key Tools Overview (Excel, SQL, BI, Python)
Stuck on this topic? Ask an Insider
Get 1:1 guidance from people who've walked this exact path — on the InsideEdge app.
