Storytelling With Data
by Cole Nussbaumer Knaflic — Read Feb 9, 2026
Most books about data visualization are really about software. This one isn't. Cole Nussbaumer Knaflic isn't teaching you Excel or Tableau. She's teaching you how to think before you ever open a tool. It's less a technical guide and more a book about communication, which is probably why it resonated with me more than I expected.
The Core Idea
The core idea is deceptively simple: most of us present exploratory analysis when we should be telling an explanatory story. Exploratory is the messy, iterative work you do to understand data—lots of charts, dead ends, hypotheses. Explanatory is what you share once you already know what it means—one clear chart, one clear point.
The problem is that most people skip the translation step. They export their exploratory work directly into a deck or dashboard, hand it to an audience, and then wonder why the room goes quiet. The audience isn't slow. They just weren't given a story. They were given raw material and asked to do the work themselves.
Start With Context, Not Charts
What I keep coming back to is her insistence on starting with context, not charts. Before you build anything, you should be able to answer: Who is the audience? What decision do they need to make? What should they do after seeing this? If you can't answer those three questions, you're probably building the wrong thing. The chart is the last step, not the first.
Before creating any chart or dashboard, Knaflic asks you to define three things: the audience, the decision they need to make, and the action that should follow. Without those answers, even accurate analysis can fail to land.
Try This
Before building any document or deck, write three lines at the top:
- Decision: What should happen?
- Insight: What did we learn?
- Evidence: What supports this?
If you can write those three lines clearly, you're ready to build. If you can't, you're not.
Choose the Right Chart (and Know What to Avoid)
Most of us default to bar charts and line charts out of habit. Knaflic gives the intuition some structure: comparisons work best as bar charts, trends as line charts, and relationships as scatter plots. Parts of a whole are trickier, and she's skeptical of pie charts for good reason—humans are genuinely bad at estimating angles.
Her stronger point is about what to avoid. 3D charts distort perception. Dual-axis charts create confusion. Decoration adds noise without adding meaning. The underlying principle is that every visual element should earn its place.
Try This
Next time you're choosing a chart type, ask whether a simpler option would work just as well. If yes, use the simpler one. Elegance in data viz is just clarity.
Preattentive Processing: Use Your Brain's Shortcut
Before your reader consciously focuses on a chart, their brain has already processed color, size, and position. That's your real estate, and most charts waste it on gridlines, legends, and decorative elements that compete with the actual insight.
The fix is usually subtractive: remove what doesn't earn its place. Use one color for data and a contrasting color for the key insight. Put the most important thing where the eye goes first. Every visual element should be asking: "Am I helping or hurting clarity?"
Try This
Take a chart you've made recently and remove every element that isn't directly supporting the main point. Gridlines, secondary labels, extra colors. See how much clearer it gets.
Structure Your Story
The storytelling framework Knaflic recommends is borrowed from narrative structure: situation, complication, resolution. The situation sets the scene. The complication introduces the tension or the finding. The resolution lands the recommendation. It works because it mirrors how people naturally process information.
Leading with the conclusion rather than the methodology is harder than it sounds. When you've spent weeks on the analysis, there's a temptation to show your work. But your audience doesn't need the journey. They need the destination.
Try This
Rewrite your next slide or section header as a complete sentence that states the insight, not the topic. Instead of "User Metrics," try "Activation dropped 18% due to onboarding friction." The difference in clarity is immediate.
The Honest Caveat
Some of the visualization guidance assumes you have design control over your outputs. In practice, templates, tools, and organizational norms often constrain you. Still, even within those limits, the underlying thinking applies everywhere: start with context, remove clutter, lead with the insight.
Worth reading if you work with data in any capacity and have ever wondered why your charts don't land the way you expected. The book won't teach you new tools, but it will change how you think about using the ones you already have.