Choosing the Right Chart for Your Data

Start with the Question, Not the Chart

The most common mistake beginners make is scrolling through the list of available charts and trying to make their data fit. The correct approach is the opposite: first, clarify the business question you want to answer, and then select the visual that answers it most effectively.

We can group most analytical questions into a few key categories. Understanding these categories provides a powerful framework for choosing visuals.

The main categories of questions you'll want your managers to think about are:

• Comparison/Ranking: How do different groups compare? Which is biggest or smallest?
• Change over Time (Trends): How has a metric evolved over days, months, or years?
• Part-to-Whole (Composition): How do different components make up a total?
• Relationship (Correlation): Is there a connection between two different measures?
• Distribution: How are the data points spread out? Where are the clusters or outliers?
• Geospatial: How does this data look on a map?

A Practical Guide for Your Workshop

For a non-technical audience of managers, providing a simple decision-making tool can be incredibly effective. A decision tree or "cheat sheet" can translate the abstract framework above into a concrete guide.

Here is an excellent example of a decision tree you can use as a handout or slide in your workshop.

Walk your workshop participants through an example using this tree:

1. Start: "Which story would you like to tell?" -> "I want to compare categories."
2. Next Question: "Do your categories have sub-categories?" -> "Yes."
3. Recommendation: -> "Use a Stacked Bar or Tree Map."

This makes the selection process less intimidating and more methodical.

Mapping Questions to Core Power BI Visuals

Now let's connect this framework to the specific visuals available in Power BI. We'll focus on the most common and effective charts that will cover 90% of a manager's needs.

The following video from Chandoo gives a fantastic, practical tour of the core Power BI visuals and when to use them.

Let's summarize how these common Power BI visuals map to our analytical questions.

1. For Comparison and Ranking

• Question: "Which country has the highest sales?" or "How do our different product lines rank by profit?"
• Best Visuals: Bar and Column Charts.
  • Column charts (vertical bars) are great when you have a time component (e.g., sales by month).
  • Bar charts (horizontal bars) are excellent when you have long category names (e.g., product names, manager names).
  • As seen in the video, these are the workhorses of data visualization for simple comparisons.

2. For Change Over Time (Trends)

• Question: "How did our website traffic change over the last year?"
• Best Visuals: Line and Area Charts.
  • Line charts are perfect for showing the shape and trend of continuous data over time.
  • Area charts are similar but fill the space under the line, emphasizing the magnitude or volume of change.
  • Crucial Tip: As the video highlights, only use line charts when your horizontal axis is a time-based dimension (dates, months, years). A line chart of sales by product name is meaningless.

3. For Part-to-Whole (Composition)

• Question: "What percentage of our total revenue comes from each business unit?"
• Best Visuals: Pie and Donut Charts.
  • These charts excel at showing how a total is divided into parts. The donut chart has the added benefit of having a hollow center where you can place a KPI card showing the total.
  • Crucial Tip: Use these only when you have a small number of categories (ideally 5 or fewer). With too many slices, they become impossible to read. A bar chart is often a better alternative for comparing many parts.

4. For Relationships (Correlation)

• Question: "Is there a relationship between the number of sales calls an employee makes and the total revenue they generate?"
• Best Visual: Scatter Plot.
  • A scatter plot places two different numeric measures on the X and Y axes. Each dot represents a data point (e.g., a salesperson). This allows you to spot patterns, clusters, and outliers instantly.

5. For Showing Detail

• Question: "I see the summary, but I need to see the exact numbers for each transaction."
• Best Visuals: Tables and Matrices.
  • Never underestimate the power of a simple table. When exact values are needed, or when you need to look up specific information, tables are essential. A good dashboard often combines high-level charts with a detailed table for users who want to dig deeper.

For a comprehensive reference, the official Microsoft documentation and the OKViz visual guide are excellent resources.

Conclusion

In this lesson, you've learned a structured approach to selecting the right visual for any analytical task. This is a critical skill to pass on to the managers in your workshop, empowering them to tell clear and compelling stories with their data.

Key Takeaways for Your Workshop:

• Start with the Question: Always clarify the business question before choosing a chart.
• Use a Framework: Teach the main categories of analysis: Comparison, Trend, Part-to-Whole, and Relationship.
• Provide Simple Tools: Use visuals like the decision tree to make the selection process easy and repeatable for non-technical users.
• Master the Core Charts: Focus on the essentials: bar/column charts for comparison, line charts for trends, and pie/donut charts for composition. These will solve most business needs.

Preview of the Next Lesson:
We've now established a solid foundation for choosing the right visual. In the next lesson, we will do a practical deep-dive into creating two visual types that are especially important for managers: we will build maps to visualize geographic data and KPI cards for performance monitoring.