AI did not end the data literacy crisis. It changed it. As machines move from providing information to interpreting it and recommending what to do, the scarce skill is becoming judgment: knowing what to trust, what to question, and what decisions are still ours to make.
Evidence does not make decisions. People interpret it, weigh it, challenge it, and make judgments. A courtroom reveals what leaders can learn about evidence, expertise, confidence, and data-informed decision-making.
Feeling sure doesn't mean you're right. Learn five simple ways to question what feels true, spot what might be influencing you, and make better decisions about what to believe.
Every chart answers one of five kinds of question. Name the question first, then the chart family follows.
Use these five as a mental filter. When you know which question you are asking, the chart family follows almost mechanically. The eight concepts below develop the thinking skills that go alongside this choice.
Essential principles for selecting and creating effective visualizations
While there are hundreds of possible chart types out there, from violin plots to Sankey diagrams, the reality is that you only need to master a core set of about 6-8 visualizations to handle 85-90% of real-world questions you'll encounter. Focus on these fundamentals before exploring more exotic chart types.
The visualization you build to understand the data yourself is not the same as the one you build to share with stakeholders. Exploration charts carry more detail to aid your analysis. Communication charts strip out anything not essential to the story you need to tell. Match the chart to the audience and purpose.
There are many times we will see data and then build a visualization before we have a clear understanding of what question we should be asking. We really should think of the question(s) beforehand.
Use: Stacked Bar Chart
Use: Line Chart
The level of measurement determines which visualizations make sense. Ordinal data (with order) works well with line charts, while nominal data (no order) can be misleading in line visualizations.
Stages have natural order
No natural order - creates false trends
Many times you'll have multiple questions about your data. It is perfectly OK to have multiple visualizations. Very rarely will you only have one question for your data. You may want to see a summary, then drill in and filter for root cause analysis.
If your question has a single answer, like "How do our sales compare this year to last year?" and the answer is just "up 12%", you can just use text. The goal of visuals is to make it easy to interpret so if the answer is a single number, no need for a chart.
Year-over-year growth
Single metric, no additional context needed
Showing trend, breakdown, or comparison
While data doesn't lie, it's important to understand the context around the data, especially when looking at trends and patterns. Don't just use a visualization to highlight something without understanding if there's another reason to explain why the data looks the way it does.
Charts can be misleading, even unintentionally. Our brains see a visual and if it has low cognitive load and makes sense, we tend to unconsciously agree with it. We need to question visualizations and apply critical thinking.
Makes small differences look dramatic
Shows true proportions
What's coming in Week 4 and beyond
In Week 4, we'll add design elements to make visualizations easier to understand: color, reference lines, clear titles, annotations, and more. We'll also explore diagnostic analytics, moving from "what happened" to "why it happened" through interactive visualizations.
