Skip to main content

Command Palette

Search for a command to run...

Analytic Approach Based on the Question Type

Updated
3 min read

When choosing an analytic approach for a problem, the type of question you’re trying to answer greatly influences the methodology. Here are five common types of questions and corresponding analytic approaches:

1. Descriptive Questions: “What is the current status?”

Approach: Descriptive Analytics

Question: "What is the current status of our sales?"

Techniques:

  • Data aggregation: Combining data from various sources into a unified view.

  • Data mining: Extracting useful information from large datasets.

  • Data visualization: Using visual tools to present data in an easily understandable format.

Examples:

  • Summarizing sales data

  • Creating dashboards

  • Generating reports

2. Diagnostic Questions: “Why did it happen?”

Approach: Diagnostic Analytics

Question: "Why did our sales decline in the last quarter?"

Techniques:

  • Drill-down: Exploring detailed data to find underlying causes.

  • Data discovery: Identifying patterns and relationships in data.

  • Correlation analysis: Assessing the relationship between different variables.

Examples:

  • Identifying root causes of sales decline

  • Analyzing customer complaints

  • Understanding failure points in a process

3. Predictive Questions: “What is likely to happen?”

Approach: Predictive Analytics

Question: "What is our sales forecast for the next year?"

Techniques:

  • Regression analysis: Predicting outcomes based on relationships between variables.

  • Time series forecasting: Predicting future values based on past trends.

  • Machine learning models: Using algorithms to predict future outcomes based on historical data.

Examples:

  • Forecasting sales

  • Predicting customer churn

  • Estimating future demand

4. Prescriptive Questions: “What should we do?”

Approach: Prescriptive Analytics

Question: "What should we do to increase website traffic?"

Techniques:

  • Optimization models: Finding the best solution from a set of alternatives.

  • Simulation: Modeling scenarios to predict outcomes.

  • Decision analysis: Evaluating and comparing different decisions.

Examples:

  • Recommending inventory levels

  • Optimizing marketing campaigns

  • Determining pricing strategies

5. Classification Questions: “Which category does this belong to?”

Approach: Classification (Supervised Learning)

Question: "Which category does this data point belong to?"

Techniques:

  • Logistic regression: Predicting the probability of a categorical outcome.

  • Decision trees: Splitting data into branches to classify it.

  • Support vector machines: Finding the best boundary to separate categories.

  • Neural networks: Using interconnected nodes to classify data.

Examples:

  • Email spam detection

  • Image classification

  • Disease diagnosis

Understanding these different types of questions and the corresponding analytic approaches can help you unlock your data's true potential.


Source: IBM Data Science Professional Certificate (Coursera)

Instructors: A special thanks to the incredible IBM instructors and the entire IBM Skills Network Team.

Disclaimer: This blog post is part of my personal learning journal where I document my progress through the IBM Data Science Professional Certificate. These articles represent my personal understanding and interpretation of the course material. They are not official course notes and are not endorsed by IBM or Coursera.