📊 Descriptive Analytics
Answers: What happened? — Summaries, dashboards, monthly reports and KPIs. Example: Monthly sales reports, traffic dashboards.
In today’s digital world, Data Analytics is the backbone of modern business decision-making. From analyzing sales trends to forecasting customer behaviour — not all analytics is the same. This guide explains the 4 key types with examples and business use-cases.
Answers: What happened? — Summaries, dashboards, monthly reports and KPIs. Example: Monthly sales reports, traffic dashboards.
Answers: Why did it happen? — Root-cause analysis, drill-downs, correlations. Example: Investigating causes of a sudden sales drop.
Answers: What is likely to happen? — Forecasting, ML models, time-series. Example: Forecasting next quarter’s revenue.
Answers: What should we do? — Decision optimisation, recommendations, simulations. Example: Recommendation engines or inventory reordering rules.
Learn how descriptive, diagnostic, predictive, and prescriptive analytics power smarter decisions across marketing, product, finance, HR and more. Read quick examples below — and see which skills will boost your career or team outcomes.
Descriptive analytics answers “What happened?” using dashboards, KPIs and summary statistics. It’s the first step teams use to verify performance before they dig deeper.
Example: A retail team uses a monthly sales dashboard to compare regions, spot anomalies and find the top 5 products that drove revenue — all classic descriptive analytics.
Use-cases below show **how each role turns analytics into action**. Click the role to jump to a short example and tools to learn.
Quick win: Use descriptive dashboards to measure campaign impact; predictive models to estimate next-month ROI.
See marketer example →Quick win: Diagnose feature drop-offs, run A/B tests and use prescriptive analytics for roadmap trade-offs.
See product example →Quick win: Forecast revenues and optimise cash flows using predictive and prescriptive models.
See finance example →Quick win: Analyze attrition reasons, design retention policies and measure inclusion metrics.
See HR example →When metrics move unexpectedly, teams need answers fast. Diagnostic analytics helps you go beyond “what happened” by identifying root causes, testing hypotheses, and finding the fixes that matter — fast.
Definition: Diagnostic analytics answers “Why did it happen?” through drill-down reports, comparisons, statistical tests and cohort analysis — the practical detective work behind data-driven improvements.
Example: An e-commerce team sees conversion fall 15% overnight. Using diagnostic analytics they segment by traffic source, device and checkout step. The team finds a mobile checkout bug introduced in a recent release — fix deployed, conversion recovers. Takeaway: Diagnostic analytics turns suspicious metrics into actionable fixes.
Pro tip: Pair diagnostic reports with a short A/B test to validate the cause before rolling out wide fixes.
Predictive analytics answers “What is likely to happen?” by training models on historical data to forecast future events — customer churn, demand spikes, sales trends — so teams can plan and act proactively.
Example: A streaming service uses logistic regression and gradient boosting on engagement metrics to predict which users will churn next month. With targeted retention emails and offers, they reduce churn by 12%—showing how predictive models convert into measurable business results.
Takeaway: Predictive analytics helps teams move from reaction to anticipation, saving cost and improving retention.
Tools: Python (scikit-learn, statsmodels), R (forecast, caret), and platforms like Prophet, Azure ML, or AWS SageMaker are commonly used for predictive pipelines.
Prescriptive analytics answers “What should we do?” by combining descriptive, diagnostic and predictive insights with optimization and simulation to recommend concrete decisions that improve outcomes.
Example: A logistics firm optimizes delivery routes by combining traffic forecasts and vehicle constraints into an integer programming model — reducing fuel costs and delivery times. Takeaway: Prescriptive analytics turns predictions into executable decisions that save time and money.
Tools & frameworks: Python (PuLP, OR-Tools), AMPL/Gurobi for optimization, SimPy for simulation, and RL frameworks like RLlib for prescriptive ML.
Explore a side-by-side comparison of the four types of data analytics — their purpose, main question, techniques, and expected outcomes. This quick guide helps you decide which analytics method fits your business challenge best.
| Type of Data Analytics | Purpose | Key Question | Techniques Used | Outcome |
|---|---|---|---|---|
| Descriptive Analytics | Summarizes historical data to understand past events and trends. | What happened? | Aggregation, visualization, reporting | KPIs, dashboards, performance summaries |
| Diagnostic Analytics | Finds the reasons behind outcomes and performance changes. | Why did it happen? | Drill-down, correlation, segmentation, root cause analysis | Explains causes of success or failure |
| Predictive Analytics | Uses past data and ML models to forecast future outcomes. | What is likely to happen? | Regression, time-series, machine learning, classification | Predicts risks, opportunities, and behavior patterns |
| Prescriptive Analytics | Recommends best possible actions and strategies. | What should we do? | Optimization, simulation, AI recommendations, reinforcement learning | Actionable plans for decision-making |
✅ Together, these four categories — Descriptive, Diagnostic, Predictive, and Prescriptive Analytics — form the complete journey of data-driven decision-making. Test your understanding →
Match your **business goal**, **data maturity**, and resources to the right analytics type — descriptive, diagnostic, predictive or prescriptive — so you turn data into fast, measurable impact.
Goal: Understand past performance.
✅ When to use: You need dashboards, monthly reports, or a performance baseline (e.g., sales by region).
Goal: Diagnose issues or anomalies.
✅ When to use: Investigating sudden KPI drops or unexplained spikes (e.g., checkout funnel fall-off).
Goal: Forecast trends or risks.
✅ When to use: Planning resources, estimating churn, or demand forecasting.
Goal: Recommend optimal actions.
✅ When to use: Optimization problems — routing, inventory reorders, personalized pricing.
Need help choosing? Try the quick quiz to match your business problem to the right analytics type.
We explored the four core analytics approaches — each one answers a different question and adds different value:
Start simple: build descriptive dashboards to create a reliable baseline → diagnose problems → layer predictive models → then use prescriptive techniques to automate better decisions.
12 real-life questions on Descriptive, Diagnostic, Predictive, and Prescriptive analytics. Select answers and submit to see instant feedback.
Diagnostic analytics explains why something happened (root-cause, segmentation, A/B testing). Predictive analytics uses historical data and models to forecast what is likely to happen next (churn, demand).
Combine **descriptive analytics** (live dashboards & alerts) with lightweight **diagnostic** checks for quick triage. For instant automated actions, use streamlined **prescriptive** systems that apply business rules or optimized recommendations.
Begin with clean historical data and a clear objective (e.g., minimize cost or maximize fill rate). Pilot a small what-if or optimization model, validate results, then scale while monitoring business metrics.
Descriptive: dashboards, summary statistics
Diagnostic: cohort & root-cause analysis, A/B testing
Predictive: regression, ARIMA, classification models
Prescriptive: Monte Carlo simulation, linear/integer programming, reinforcement learning
The four main types are Descriptive (what happened), Diagnostic (why it happened), Predictive (what will likely happen) and Prescriptive (what should be done).
इस वीडियो में हमने चार मुख्य प्रकार के डेटा एनालिटिक्स – Descriptive, Diagnostic, Predictive, और Prescriptive – को सरल हिंदी में समझाया है। यह शुरुआत करने वालों के लिए एक बेहतरीन विज़ुअल गाइड है।
Expert-led training in data analysis, visualization, and machine learning — with hands-on projects and real-world case studies to build your portfolio.
Address: 316/336, Park Rd, Laxman Chowk, Dehradun, Uttarakhand 248001
Phone: 094117 78145
Rated 4.8/5 by 1,200+ learners
•Hands-on projects • Job-oriented curriculum • Mentor support
Data Analyst @ Dehradun Startup
“The capstone on sales forecasting landed me my first analytics role. The mentors reviewed my dashboards line-by-line.”
MBA → Business Analyst
“From Excel to Python + SQL in 8 weeks. The interview prep and case studies were spot-on for real roles.”
Fresher → Junior Analyst
“Loved the live projects—built a churn model and a Power BI dashboard that I now show in interviews.”
Ops → Data Ops
“Weekend classes + recorded sessions made it easy to switch. The placement team actually connects you to recruiters.”