Use this free scatter plot generator to compare two numeric variables, see clusters, and inspect unusual observations. The same scatter plot maker — sometimes written as one word, scatterplot generator — works from CSV, Excel, or pasted cells. Choose X and Y fields and optionally color points by group.
CSV, TSV, XLS/XLSX or pasted cells. Your spreadsheet contents stay on this device.
Save or reopen data and style
Download a local file with your table, title, theme, source and notes. Reopening detects fields again; review chart-specific settings. No data is uploaded.
Start with your table or try the sample
Numeric X + numeric Y + optional group
No account or server-side file processing.
Paste data from Google Sheets
In Google Sheets, select the cells with their column headers and copy them. Choose Paste data above, paste the table, and choose Use pasted data. Copy the cell range rather than a chart image or a spreadsheet sharing URL.
Copy the paired X and Y numeric columns with their headers, plus a group column if needed. Each row must describe the same observation across both measurements. Do not sort either measurement column on its own. Review X field, Y field, and Group by after pasting.
This is a separate chart made from the copied values; later changes in Sheets do not update it. The table is processed locally in your browser.
Input requirements
Give the chart the right data
Choose two different numeric columns. Each row is one observation. Optionally select a third column with up to 12 groups. Negative and repeated coordinates are supported. Empty selected cells must be corrected; observations are not dropped.
An XY graph maker and an XY scatter chart describe the same figure: each observation supplies both an X and a Y value, and every row becomes a point. If you searched for an XY graph maker, choose two numeric columns below. The axes represent two measurements, rather than a category and its count.
Maximum import: 10 MB, 2,000 rows including the header, 100 columns, and 20,000 cells. This tool accepts zero or absolute numeric values between 10⁻⁹ and 10¹²; rescale extreme values first. Excel imports read the first worksheet.
Choose deliberately
When to use a scatter plot
A scatter chart generator is useful for advertising spend versus sign-ups, price versus demand, or two measurements from the same observations. Each row becomes one point you can hover to inspect its original values.
When to choose another chart
A pattern does not establish causation. Overlapping observations can hide density; use smaller, more transparent points or explore one variable with the histogram maker. For change over an ordered sequence, choose the line graph maker. An optional overall linear trend line summarizes the displayed points. It does not establish causation or statistical significance.
Scatter plot vs line graph: which fits your data?
A line graph connects observations in a meaningful order, such as dates, to show how a measure changes. A scatter plot leaves rows unconnected so you can examine how two numeric measurements vary together. If the first column contains dates, use the line graph maker and its date mode. If each row pairs two measurements from the same campaign, person, or experiment, use a scatter plot.
For example, monthly sign-ups belong on a line graph when the question is how they changed over time. Spend and sign-ups for individual campaigns belong on a scatter plot when the question is whether larger spend tends to occur with more sign-ups.
Common mistakes when reading a scatter plot
Overlapping points hide observations. Two rows with the same coordinates occupy one visible position. Reduce point size or opacity and check the data table before judging how many observations you have.
An outlier is not automatically an error. Check its source row and context before excluding it. A valid unusual point may change the apparent pattern.
Association is not causation. Other variables can affect both axes, and the direction of influence may be unclear. Describe the visible pattern; make causal claims only with evidence beyond the chart.
How do I add a trend line to a scatter plot?
Import two numeric columns, review X and Y, and enable Show linear trend line. The dashed line is an overall least-squares fit across all selected groups. Its equation and R² are included in PNG and SVG exports. R² describes the fit to these observations, not causation or statistical significance. The slope depends on measurement units; a steeper line alone does not mean a stronger relationship.
With identical X values the fit is unavailable. With constant Y values R² is not applicable. Review outliers and nonlinear patterns before interpreting a straight line.
From table to figure
Three steps to a useful chart
Import and review
Upload a spreadsheet, paste cells, or load fictional sample data. Review the automatically selected fields and correct them if necessary.
Read and refine
Inspect the values and any data warnings. Choose a publication style, canvas, title, and source note.
Export your result
Download editable SVG or PNG at 1×, 2×, or 3×. The original spreadsheet contents stay in your browser.
Before you create your chart
Does every row produce a point?
Yes. Each row produces one point, including repeated coordinates. Identical positions overlap; the accessible data table preserves every observation.
Can each axis have a different unit?
Yes. X and Y are validated separately, so one may contain percentages while the other contains ordinary numbers. Mixed formats within one column are rejected.
How do I color points by channel?
Select a separate field in Group by. The tool allows at most 12 groups; choose None to use a single color.
How many points can I plot?
An import can contain up to 2,000 rows including the header, so at most 1,999 observations, subject to the 20,000-cell and 10 MB limits. A large number of overlapping points may still be difficult to read.
Can I add a trend line?
Yes. Enable Show linear trend line for an overall least-squares fit across all points and groups. The chart includes its equation and R² in PNG and SVG exports. Identical X values cannot define a fit; constant Y values have no defined R². A trend does not prove causation.
What is the difference between a scatter plot and a scatter chart?
They are two names for the same chart: each point combines an X and Y measurement from one observation.
Can I plot dates on the X axis?
The scatter plot requires numeric X values, so date strings cannot be used directly. For a time series, use the line chart's date mode; convert dates to numeric values only if that scale answers your question.
Does Charttoolbox upload my data?
Spreadsheet contents are parsed, calculated, and rendered locally. Optional Google Analytics loads only after consent and does not receive filenames, spreadsheet contents, or generated chart data.
Explore related chart tools
Use a tool that matches the question you are asking.
See what each point represents, how to describe association, and why it does not prove causation in What is a scatter plot?
Help us improve Charttoolbox
With your permission, we use Google Analytics to understand visits and feature usage. We never send file names,
spreadsheet contents, cell values, or chart data. Privacy details