Scatter Plot Maker: Create Scatter Plots Online From CSV Data
Need to explore the relationship between two numeric variables? Paste a CSV, map the X and Y fields, then add color or point size to see clusters, outliers, and correlation patterns more clearly.
- Best for correlation checks, experiment results, performance benchmarking, and outlier detection
- Works with tidy CSV or TSV data where each row is one observation
- Exports the finished chart as PNG or SVG after you configure the view
What a scatter plot is and why it matters
A scatter plot places one numeric variable on the X axis and another numeric variable on the Y axis so you can see how the two move together.
This is one of the fastest ways to answer questions like:
- do higher X values usually come with higher Y values
- are there natural clusters in the data
- are there outliers that deserve a second look
- does a third variable change the pattern when shown as color or point size
If the main story is change over time, use a Line Chart Maker. If you need category comparisons instead of numeric relationships, use a Bar Chart Maker.
Which scatter plot setup should you choose?
| Setup | Best for | Example |
|---|---|---|
| Basic scatter plot | Relationship between two numeric variables | ad_spend vs revenue |
| Colored scatter plot | Comparing groups within the same relationship | study_hours vs score by class |
| Sized scatter plot | Adding a third numeric measure | price vs sales sized by margin |
| If both axes are numeric and each row is one observation, you already have the right shape for a scatter plot. |
Format your data for this scatter plot maker
The cleanest input is a tidy table with:
- one numeric X column
- one numeric Y column
- optional color or grouping column
- optional numeric size column
Example:
study_hours,test_score,group,sleep_hours
2,58,Group A,6.1
3,62,Group A,6.4
4,67,Group A,6.8Each row should represent one point. If you need a chart for ordered trends rather than variable relationships, switch to Line Chart Maker, Area Chart Maker, Bubble Chart Maker, or CSV to Chart.
Common scatter plot mistakes
- Do not use text columns for the X or Y axes. Scatter plots need numbers on both axes.
- Do not infer causation from a visual correlation alone.
- Do not overload the chart with too many categories if readers only need one comparison.
- Do not use point size as decoration. It should represent a real numeric field or stay fixed.
If the third numeric variable is essential to the chart rather than optional, switch to the dedicated Bubble Chart Maker.
FAQ
What is the best data format for a scatter plot maker?
The best format is tidy CSV or TSV with one row per observation and numeric columns for both the X and Y axes.
Can I color points by category?
Yes. Use a categorical column such as group, region, or product as the color field to compare clusters and segment patterns.
What if I want point size to represent another metric?
Use a numeric size field. The maker can scale point area by that column so you can compare three variables in one chart.