← IndexEntry № 068·data

Choose the Right Chart Type for Your Data

Get a chart type recommendation with rationale, alternatives, and common pitfalls for your data.

Optimized for
ChatGPTClaudeGemini
§ When to use this

This prompt turns a vague 'what chart should I use?' into a reasoned recommendation: it asks the model to match a chart type to both your analytical goal and the shape of your data, then justify it, offer a fallback, warn about pitfalls, and sketch the axis mapping. Reach for it before building a dashboard or slide when the wrong chart would quietly mislead your audience. Forcing it to name a second option and common mistakes is what separates this from a one-word answer — the trade-offs are usually where the real decision lives.

§ The Prompt— fill in the fields, then copy or open in a tool
§ Customize0/2 fields filled
your prompt — fill the fields above
You are a data visualization expert. Help me pick the right chart type. I want to show [WHAT YOU WANT TO SHOW: e.g. trend over time, part-to-whole, comparison, distribution, correlation] using data that has [DESCRIBE DATA: e.g. 12 months of revenue, 5 categories, two continuous variables]. Recommend: (1) the ideal chart type and why, (2) a second option and when to use it instead, (3) common mistakes to avoid with this data type, (4) sketch the axis labels and data mapping in text form.
Open with your prompt →ChatGPTClaudeSends your filled-in prompt straight into a new chat.
§ Example Output

What you can expect back

1. Ideal: Line chart.
A continuous line is the clearest way to read direction and momentum across 24 evenly-spaced time points.

2. Second option: Bar chart.
Use bars instead if each month is a discrete reporting period you want compared individually rather than read as a flow.

3. Common mistakes:
- Truncating the y-axis (not starting at 0) exaggerates the trend.
- Adding a second y-axis to overlay another metric — it distorts perceived correlation.
- Too many gridlines burying the line.

4. Axis mapping:
- X-axis: Month (Jan 2024 -> Dec 2025), evenly spaced.
- Y-axis: Revenue ($), starting at 0.
- Series: single line, one point per month, points optional.

Illustrative example — your results will vary by tool and inputs.

§ Pro Tips

Get sharper results

  • 01Tell it who the audience is — an exec deck and an analyst report justify different chart choices, and the model will weigh simplicity vs. detail accordingly.
  • 02Mention the medium (static slide, interactive dashboard, mobile) because some chart types that work interactively fall apart as a printed image.
  • 03If you're tempted by a pie chart, ask it specifically when a pie is and isn't appropriate — it'll usually steer you to bars unless you have very few parts of a clear whole.
  • 04Ask for the exact encoding (color, size, position) it recommends, not just the chart name, so you don't accidentally encode a category as a continuous gradient.
  • 05Push it to flag when your data actually needs two charts instead of one overloaded combo chart.
§ Variations

Adapt it for your case

Tool-specific build steps

Add 'I'm using Excel / Tableau / matplotlib' and ask for the concrete steps or code to build the recommended chart.

Accessibility-first

Ask it to recommend a chart that remains readable for colorblind viewers and to suggest a colorblind-safe palette.

Story-driven choice

Describe the one takeaway you need the audience to leave with and ask which chart most directly drives that conclusion.

Use For — Tasks
Tags#data-viz#charts#analytics
§ FAQ

Common questions

It recommended a chart I think is wrong — now what?

Tell it your objection and the constraint it missed (audience, medium, data quirk). Chart choice is contextual, and adding the missing context usually changes the answer.

Can it help with more exotic charts like Sankey or heatmaps?

Yes — describe your data honestly and it will suggest specialized types when they fit, plus warn you when a simpler chart communicates better.

Will it actually generate the chart?

This prompt sketches the mapping in text. To produce the visual, add your tool and ask for the code or step-by-step build, or use a tool with chart-rendering capabilities.

§ Related Entries

You may also need