Welcome to your first lesson. We'll begin by building a strong foundation for your journey into data analytics. Your goal is to lead your team effectively, and a core part of that is understanding the language they speak and the value of the work they do. This lesson is designed to give you that essential vocabulary.
We will focus on the four fundamental types of data analytics. By the end of this lesson, you will be able to distinguish between descriptive, diagnostic, predictive, and prescriptive analytics. This framework will help you categorize the projects your team undertakes, understand the questions they are trying to answer, and appreciate the different levels of complexity and business impact involved.
The Four-Fold Path of Analytics
At its core, data analytics is about asking questions and using data to find answers. These questions typically fall into four categories, each representing a type of analysis that builds on the last, moving from understanding the past to shaping the future.
Let's start with a high-level overview. The following infographic provides a great summary of the four types of data analytics, the main question each one answers, and the typical steps involved in each process.
As you can see, the journey often starts with a simple question: "What happened?" and can progress all the way to "What should we do?". Let's explore each of these stages in more detail.
1. Descriptive Analytics: What Happened?
This is the most common and foundational type of analytics. Its sole purpose is to summarize historical data to provide a clear picture of the past. Think of it as creating a concise report of events.
- Purpose: To track Key Performance Indicators (KPIs), monitor business health, and identify trends.
- Common Outputs: Business intelligence (BI) dashboards, sales reports, website traffic summaries.
- Example: Your team uses Power BI to create a dashboard showing last quarter's sales figures, broken down by region and product. This dashboard is a classic example of descriptive analytics. It doesn't explain why sales in the North region were highest, but it clearly states that they were.
Given your background in web development, you can think of this as the logging and monitoring layer. You collect data on server response times, error rates, and user activity. Descriptive analytics organizes and presents this data, for instance, in a graph showing the average response time over the last 24 hours.
2. Diagnostic Analytics: Why Did It Happen?
Once you know what happened, the natural next question is why. Diagnostic analytics is the investigative or "detective" phase. It involves drilling down into the data to find the root causes of the outcomes identified by descriptive analytics.
- Purpose: To uncover the factors and relationships that explain a particular trend or event.
- Common Techniques: Drill-down analysis, data segmentation, correlation analysis.
- Example: Using the sales dashboard from before, a stakeholder asks why sales for a particular product suddenly dropped in July. Your analyst team would perform diagnostic analysis. They might segment the sales data by marketing campaigns, competitor activity, and customer demographics. They might discover that a key competitor launched a major discount campaign that month, which is the likely cause of the sales dip.
3. Predictive Analytics: What Will Happen?
This type of analytics moves from looking at the past to forecasting the future. It uses historical data, statistical algorithms, and machine learning techniques to estimate the likelihood of a future outcome.
- Purpose: To anticipate future trends, identify risks, and find opportunities. It's about making informed guesses, not declaring certainties.
- Common Outputs: Sales forecasts, customer churn scores, demand predictions.
- Example: An e-commerce company wants to identify customers who are likely to stop purchasing. A data analyst or data scientist could build a predictive model based on historical customer behavior (e.g., time since last purchase, frequency of visits, items in cart). The model would then assign a "churn risk" score to each current customer, allowing the marketing team to proactively target high-risk customers with retention offers.
4. Prescriptive Analytics: What Should We Do?
This is the most advanced form of analytics. It goes beyond predicting an outcome to recommend specific actions to take to achieve a desired goal or to mitigate a future risk.
- Purpose: To provide optimal recommendations and guide decision-making by simulating the potential outcomes of different choices.
- Common Techniques: Optimization algorithms, A/B testing frameworks, AI-driven decision rules.
- Example: An airline wants to maximize revenue for a specific flight. A prescriptive analytics model would combine predictive data (demand forecasts, competitor pricing) with business constraints (number of seats, fuel costs). It would then not only predict demand but also recommend the optimal ticket price at any given moment to maximize profitability.
Bringing It All Together
The progression from descriptive to prescriptive analytics represents an increase in both business value and complexity. To see these concepts illustrated with a single, clear analogy, watch the following short video.
Descriptive vs Diagnostic vs Predictive vs Prescriptive Analytics: What's the Difference?
This video from Maven Analytics uses a relatable doctor's visit scenario to explain the differences between the four analytics types.
Pay close attention to how the doctor's statements evolve from a simple descriptive fact to a prescriptive action plan from the example. Also, note the important clarification at the beginning that these types are often used in tandem rather than as a strict linear progression you must "graduate" from (the clarification).
To deepen your understanding and see more business-focused examples, the following article provides a thorough breakdown.
The 4 Types of Data Analytics Explained + Examples
This article from Domo offers detailed explanations and real-world use cases for each type of analytics. It provides a solid foundation for the business context in which these techniques are applied.
Please read the following sections: Start with the comparison table under the heading How the 4 types of data analytics compare. This gives a very concise overview. Then, read the sections defining descriptive analytics, diagnostic analytics, predictive analytics, and prescriptive analytics. Focus on the "How it works" and "Examples" subsections for each.
As you read, think about the work your own team does. Can you map their projects to these categories? A project to build a daily performance dashboard is descriptive. An investigation into a sudden spike in user unsubscribes is diagnostic. A model to forecast server load is predictive. This framework is your first tool for structuring your understanding of your team's activities.
Conclusion
In this lesson, we have established the foundational vocabulary for discussing data analysis. You can now distinguish between the four main types of analytics:
- Descriptive Analytics: What happened? (Summarizing the past)
- Diagnostic Analytics: Why did it happen? (Investigating causes)
- Predictive Analytics: What will happen? (Forecasting the future)
- Prescriptive Analytics: What should we do? (Recommending actions)
Understanding this progression from hindsight to foresight is the first step toward effectively managing your data team. It allows you to classify their work, appreciate the relative complexity of their tasks, and better articulate the business value of their efforts to other stakeholders.
In our next lesson, we will dive a level deeper. Now that you understand the broad types of analysis, we will define the key terminology used within these analyses, such as KPIs, metrics, dimensions, and measures. These are the fundamental building blocks of the reports and dashboards your team creates.