A clean dataset and a solid data model can only take you so far in Power BI. If your visuals don’t clearly communicate the message, all that work might go unnoticed. Visuals are the lens through which your audience sees the data and choosing the right one is key to turning information into insight.
In my earlier article, From Data to Decisions: How to Build High-Impact Power BI Reports, I emphasised that effective reports are more than just functional, they’re purposeful, engaging, and built around the user. This article builds on that foundation by exploring how to select the right visuals to clearly and truthfully communicate your data.
We’ll focus on Power BI’s core built-in visuals and explore how to apply them effectively. But first, we need to understand what makes a visual good in the first place.
Key Takeaways
- Choosing the right visual is essential for turning raw data into clear, actionable insight in Power BI.
- Edward Tufte’s principles, such as reducing chartjunk and increasing the data-ink ratio offer timeless guidance for designing meaningful visuals.
- Power BI’s core visuals (bar, line, scatter, maps, cards, etc.) each serve a distinct purpose and let the message drive your choice.
- Even simple charts like bar charts can be made significantly more effective by applying thoughtful design principles.
- Ongoing learning through trusted resources and hands-on practice is key to developing strong visual storytelling skills.
The Foundations of Good Data Visualisation: Lessons from Edward R. Tufte
Before diving into specific visuals, it’s worth asking: What actually makes a chart effective? This question sits at the heart of Edward R. Tufte’s influential book, The Visual Display of Quantitative Information.
Tufte’s work is foundational in the world of data visualisation. Although published decades ago, his principles still offer valuable guidance for anyone designing data displays today, especially in interactive tools like Power BI.
His focus is on clarity, precision, and integrity. Has principles are helping us move beyond just making charts to actually communicating meaning.
Data-Ink Ratio
One of Tufte’s best-known concepts is the data-ink ratio. The idea that the majority of what you see in a visual should represent actual data, not decoration. In practical terms, this means stripping away nonessential elements like borders, shadows, and background patterns. Less clutter means more focus on the data.
“Above all else show the data.” — Edward R. Tufte
Chartjunk and Visual Noise
Tufte coined the term chartjunk to describe unnecessary visual elements that distract from the message. This includes things like 3D effects, ornate icons, and overly complex colour schemes. These elements may look impressive at first glance, but they dilute the meaning and increase cognitive load.
Integrity of the Data
Perhaps Tufte’s most critical principle is that data should be represented honestly. Visuals must avoid manipulation, like cutting off axes to exaggerate differences, using distorted proportions, or selectively presenting data. The goal is to inform, not to persuade or mislead.
Small Multiples and Comparisons
Tufte often advocates for small multiples, which are a series of simple, consistent charts shown side-by-side to make comparisons easier. This approach makes patterns more visible without relying on a single overloaded chart. It’s a powerful way to show differences over time, across regions, or between categories.
Clarity Through Simplicity
Good visual design is not about flash but clear thinking. Every element of a chart should serve a purpose. When design choices are intentional and minimal, the message becomes obvious and memorable.
Understanding the Core Power BI Visuals
Now that we’ve set the stage with Tufte’s principles, we can look at how to apply them using Power BI’s built-in visual types. Instead of choosing visuals based on habit or appearance, we’ll focus on what the data is trying to say and how best to let it speak.
Here’s a breakdown of Power BI’s core visuals, grouped by intent and supported by practical tips for use.
1. Comparison: Highlighting Differences Between Categories
- Clustered Column Chart
- Clustered Bar Chart
- Stacked Column/Bar Chart
Use these visuals to compare values across categories, such as sales by region or spend by department.
Tips:
- Sort by value or a logical sequence, not alphabetically.
- Avoid clutter like shadows or unnecessary gridlines.
- Keep the focus on clear differences, not design flourishes.
2. Trends Over Time: Tracking Change
- Line Chart
- Area Chart
- Combo Chart (Line and Column)
These visuals show how data changes over time, which are perfect for metrics like revenue, customer growth, or performance trends.
Tips:
- Use line charts for clean, straightforward timelines.
- Use area charts only when the cumulative value is part of the message.
- Avoid dual axes unless absolutely necessary and label clearly if used.
3. Part-to-Whole Relationships
- Pie Chart
- Donut Chart
- 100% Stacked Column/Bar Chart
These visuals show how individual parts contribute to a whole. Use them with caution and only when the number of categories is small and the message is simple.
Tips:
- Don’t rely on these for precise comparisons.
- If exact values matter, bar charts are often a better alternative.
- Pie and donut charts are visually appealing but often fail Tufte’s test of clarity and precision.
4. Distribution and Frequency
- Column/Bar Charts with Binning
- Scatter Chart
Use these visuals to show how values are distributed across a range. They’re ideal for identifying clusters, gaps, or outliers.
Tips:
- Keep bin sizes consistent and intuitive.
Label axes clearly to avoid confusion.
In scatter charts, avoid visual overload and keep points readable.
5. Relationships Between Numeric Values
- Scatter Chart
This is the go-to visual for showing correlation or relationships between two quantitative fields, which is suited for data such as advertising spend versus sales.
Tips:
- Add trend lines for context if appropriate.
- Don’t overcrowd the chart, use filters or highlights to guide the story.
Geographical Data
- Map
- Filled Map
Geographic visuals are powerful when location is central to the insight, this is the case for scenarios like store performance by state or customer reach by country.
Tips:
- Avoid unnecessary fills or gradients that make interpretation harder.
- Ensure your location data is clean and unambiguous.
Tables and Cards: Detailed or Summary Values
- Table
- Matrix
- Card
- Multi-row Card
Sometimes, a visual isn’t necessary and what you really need is the number. Use tables and cards for detailed information or high-level KPIs.
Tips:
- Use sparingly to avoid clutter.
- Combine with visuals to add context when needed.
How to Choose the Right Visual
Here’s a quick reference to help match your message with the most effective visual:
| Insight Type | Recommended Visuals |
| Comparison | Column, bar, stacked bar |
| Trend over time | Line, area, combo chart |
| Part-to-whole | Donut, 100% stacked bar |
| Correlation | Scatter chart |
| Spatial distribution | Map, filled map |
| High-level metrics | Card, multi-row card |
The key is to let the data and the purpose guide the choice. And when in doubt, remember Tufte’s advice: eliminate anything that doesn’t serve the message.
Applying Tufte’s Principles: A Real-World Bar Chart Example
To bring Tufte’s concepts into practice, let’s apply them to a common scenario using a standard Power BI bar chart.
Scenario: Visualising Monthly Sales by Product Category
You’re creating a visual to compare monthly sales across product categories in March. You choose a clustered bar chart which you think is great for direct comparison.
But how you design that chart makes all the difference. Here’s how Tufte’s principles come into play:
1. Increase the Data-Ink Ratio
- Before: Thick gridlines, bold axis labels, multiple colours.
- After: Remove gridlines or make them subtle, simplify axis formatting, and limit labels to what’s useful.
2. Eliminate Chartjunk
- Avoid 3D effects, background gradients, or overuse of colour.
- Use consistent, neutral tones unless highlighting a specific category.
- Strip out shadows, borders, and redundant legends.
3. Maintain Data Integrity
- Always start your axis at zero to avoid misleading the viewer.
- Avoid distortions caused by squeezed or exaggerated scales.
- Ensure that the order and grouping of categories don’t imply meaning where there is none.
4. Encourage Comparison Through Alignment
- Sort categories by value, not alphabetically.
- Align bars on a shared baseline for easy visual scanning.
- Use side-by-side small multiples if you need to show breakdowns across different months or regions.
5. Emphasise the Story, Not the Chart
The title should guide interpretation, not just describe content. For example:
- Instead of: “Sales by Product Category”
- Use: “Electronics Leads March Sales Across All Regions”
This helps your audience understand the takeaway at a glance.
Before vs. After: Summary
| Design Element | Over designed Chart | Tufte-Inspired Chart |
| Gridlines | Bold and numerous | Subtle or removed |
| Colour | Random or overused | Purposeful and minimal |
| Labels | Everywhere, overlapping | Focused and legible |
| Chart Title | Generic | Insight-driven |
| Axis | Truncated or inconsistent | Starts at zero, consistent formatting |
From Good Visuals to Great Decisions
Great visuals don’t just present data, they help your audience understand it. Power BI gives you the tools, but the responsibility for visual clarity and integrity lies with you.
By combining Power BI’s core visuals with the timeless guidance of The Visual Display of Quantitative Information, and the practical strategies from From Data to Decisions, you can design reports that are not only effective but insightful, ethical, and impactful.
Continue Building Your Visualisation Skills
If you’re ready to deepen your understanding of visual design and Power BI best practices, here are some curated resources to guide your learning that are centred on the principles discussed in this article.
1. Microsoft Learn – Power BI Modules
Microsoft offers free, interactive training that walks you through every part of the Power BI ecosystem, including report design and visualisation choices.
Recommended modules:
Create and customize Power BI visuals
Design effective reports for data analysis
2. Books to Deepen Your Understanding
For a deeper dive into the why behind great data visuals, these books are essential reading. They pair theory with real-world application and are highly relevant to Power BI users:
The Visual Display of Quantitative Information by Edward R. Tufte
Storytelling with Data by Cole Nussbaumer Knaflic
Information Dashboard Design by Stephen Few
3. Practice with Sample Datasets
Practice is one of the best ways to build intuition around chart selection and design. Start with Microsoft’s built-in sample datasets or download open datasets from:
Challenge yourself to:
- Redesign a cluttered visual using Tufte’s principles
- Test different visuals to communicate the same message
- Create a personal gallery of “before and after” visual transformations
By combining Power BI’s core visuals with timeless design principles and hands-on practice, you’ll move from simply building reports to crafting data stories that resonate.
Data Empowered World 
Leave a comment