Scatter Plot in Excel With 3 Variables: How to Make One

How-to basics · Updated November 21, 2026

To show three variables in an Excel scatter plot, use a Bubble chart: plot the first variable on the X axis, the second on the Y axis, and encode the third as bubble size. In Excel for Microsoft 365 on Windows, select the three data columns, then choose Insert > Insert Scatter (X, Y) or Bubble Chart > Bubble.

What a scatter plot with three variables means

A standard scatter plot positions each observation using two numerical values: one determines its horizontal location and the other its vertical location. To include a third numerical value without adding another axis, a bubble chart makes each point a circle whose size represents that value. For example, a chart could show a city’s population on the X axis, average income on the Y axis, and its number of annual visitors as bubble size.

This is still a two-dimensional chart. Bubble size is an additional visual cue, not a third spatial coordinate, so it is not the same as a 3-D scatter plot. Excel’s regular XY scatter chart does not add a third numerical axis; a Bubble chart is the usual Excel choice when you want to compare three numeric measures in one view.

Bubble charts work best when the third variable is meaningful as a quantity and bubble sizes can be compared at a glance. They are less effective when points overlap heavily or when the third variable has many similar values. A chart with three variables can show more than a basic scatter plot, but it also asks the reader to decode more visual information.

Arrange and check your data

Put each observation on one row and each variable in its own column. Use a clear header row, and keep the measurement for each observation together across the row. For example:

  • A: Product or location name
  • B: Advertising spend, to use as the X variable
  • C: Sales, to use as the Y variable
  • D: Number of customers, to use as bubble size

The name column is useful for identifying points, but the three plotted columns should contain numeric values. Remove explanatory text, subtotals, and blank rows from the middle of the data. Check that Excel recognizes numbers as numbers rather than text; values imported with unusual separators or leading apostrophes may not plot as expected. If some rows have missing measurements, decide whether to exclude those observations or correct the source rather than filling gaps with arbitrary numbers.

Bubble sizes should be positive numeric values that have a sensible interpretation. A bubble chart is not a good way to plot a category such as “North,” “South,” or “East” as the third variable. Use a separate series, labels, or another chart instead. Also consider the range: one unusually large size value can make the other bubbles appear tiny.

Create a Bubble chart in Excel

With the data in adjacent columns, try this workflow in Excel for Microsoft 365 on Windows:

  1. Select the three numeric columns, including their headers. If names are in a separate column and you want to use them as labels later, you can select those too, but keep track of which columns contain the plotted values.
  2. Choose Insert > Insert Scatter (X, Y) or Bubble Chart > Bubble. Choose the ordinary Bubble option for a flat chart; a 3-D Bubble option, if offered, changes the chart’s appearance and is not a true third measurement axis.
  3. Inspect the result before formatting. Check whether the horizontal axis uses the intended X values, the vertical axis uses the intended Y values, and the circles represent the intended size values.
  4. If Excel has assigned the columns incorrectly, use Chart Design > Select Data and edit the series, as described in the next section.

Excel may infer the series layout from the selected range, and the result can depend on how the range is arranged. If the chart does not show the expected points or bubble sizes, do not assume the source values are wrong: confirm the series ranges explicitly. On Excel for Mac, the chart commands are also on the Insert tab, though the ribbon layout and chart-type menu can differ slightly. Excel for the web has chart tools too, but some formatting and selection controls differ; use the desktop app if you need precise series editing.

Set the X, Y, and bubble-size ranges

Each bubble series needs three ranges: X values, Y values, and bubble size values. To check or correct them, select the chart and open Chart Design > Select Data. In the Select Data Source dialog, select the series and choose Edit, or choose Add to create a series. Set the series name from a header cell if useful, then assign the X, Y, and bubble-size ranges to the matching columns. The three ranges should cover the same number of observation rows and should not include the header cells as data.

For the example above, advertising spend belongs in the X values field, sales in the Y values field, and customer count in the bubble size field. If sales and spend appear on the wrong axes, swap the X and Y ranges; the chart will still contain the same observations, but its interpretation changes. If circles all look the same size, check that the bubble-size range contains varying numeric values and that it was assigned as bubble size rather than included as another series.

After correcting the series, select OK and review several individual points against their source rows. This simple check can catch a shifted range, such as using rows 2–20 for X values but rows 3–21 for Y values. A chart can look plausible even when its ranges do not line up, so verify the mapping before drawing conclusions.

Show a fourth variable with color or labels

A fourth variable can sometimes be added through color, but Excel does not automatically assign a different color to each point in one bubble series based on a category column. A practical approach is to split the observations into separate series by group—for example, one series for each product type—while using the same X, Y, and bubble-size measures in each series. Add or edit each series through Chart Design > Select Data, then format the series colors so the groups are distinguishable.

Keep the number of groups manageable. Too many colors make the chart hard to read, and overlapping bubbles from different groups can hide one another. A legend can identify series, but it does not explain every point individually. If the group names matter for individual observations, use data labels where appropriate or provide a companion table.

Excel can also display labels from cells. Select a data series, use Chart Design > Add Chart Element > Data Labels, and open the label-formatting options to select labels from cells if that option is available in your build. Choose a short, identifiable label range. Showing long names for every bubble can make the chart unreadable, so label only key points when a small subset is the focus.

Format the chart so the values are readable

Start with an informative chart title and axis titles that include units. “Advertising spend (thousands of dollars)” is more useful than “Spend,” especially if the chart will be read outside its original workbook. Add axis titles using Chart Design > Add Chart Element > Axis Titles. Format axes to use a sensible number format and range; an axis that starts at zero is not always required for a scatter plot, but the displayed range should not mislead readers about the size of differences.

Bubble size is especially easy to misread. Excel scales the displayed circles to fit the chart, so readers should treat size as a visual comparison rather than a precise substitute for reading the underlying number. If the bubbles obscure one another, select the series and adjust transparency or the bubble-size scale in the Format Data Series pane. You can also reduce the number of plotted observations or use smaller markers. Avoid using a 3-D visual effect to suggest depth: perspective can distort apparent sizes and positions.

Use a restrained color palette and include a legend when multiple series are present. A color should have one consistent meaning throughout the chart. If the chart is being printed or viewed by people with color-vision differences, do not rely on color alone to distinguish groups; consider clear labels or other supporting context.

Read the chart without overclaiming

Each bubble’s center gives its X and Y position, while its size offers a visual comparison of the third measure. Look for patterns such as a rising relationship between X and Y, clusters, or points that differ from the general pattern. Then check whether bubble size seems associated with those patterns. For example, higher sales at higher advertising spend do not, by themselves, prove that advertising caused the sales increase. Other factors may explain the relationship, and the chart is descriptive rather than a causal analysis.

Overplotting is a common limitation: when observations share similar X and Y values, their bubbles may cover one another. Smaller bubbles, transparency, filtering, or separate charts can help, but each has trade-offs. Transparency makes overlaps more visible but can make colors blend. Splitting groups into separate charts can improve clarity but makes direct comparison less immediate. If exact values or many observations matter, a table or a different visualization may be more suitable.

Also check whether the variables have appropriate units and scales. A wide range on one axis can compress the majority of points into a small area. A logarithmic axis can help with some positively valued, strongly skewed data, but it changes how distances are interpreted and cannot represent zero or negative values in the usual way. Use it only when the measurement and intended interpretation support it, and label the axis clearly.

When to use another chart

Choose a Bubble chart when the third variable is numeric, varies meaningfully across observations, and the audience benefits from seeing it alongside two numerical relationships. If the third measure is a category, use separate colored series or a different chart design. If the goal is to compare exact values across many items, a table, bar chart, or dot plot may be clearer than circles, which are harder to compare precisely than aligned positions or lengths.

For a time series, a line chart may better show changes over time, with a separate panel or color encoding for another measure. For many groups or many data points, small-multiple charts can reduce clutter. There is no requirement to place all three variables in one chart; separating the view can make the relationships easier to explain and less likely to be misread.

Frequently Asked Questions

Can a regular Excel scatter plot show three variables?

A basic XY scatter plot uses two numerical axes. To display a third numeric value in the same chart, use a Bubble chart so that the third value controls bubble size. This adds a visual encoding, not a third coordinate axis.

How do I change which column controls bubble size?

Select the chart and choose Chart Design > Select Data. Edit the series and assign the intended numeric range to the bubble-size field. Confirm that it covers the same observation rows as the X and Y ranges.

Can I use a text category as the third variable?

Not as bubble size. Use the category to divide observations into separate series, each with its own color, or label the points. A text value does not provide a meaningful numeric bubble-size measurement.

Why are some bubbles missing or hard to see?

Check for blank, nonnumeric, zero, or negative size entries, and confirm the correct range is assigned. Very small bubbles can be difficult to see, while large bubbles can cover nearby points. Review the source values and adjust the series scale or chart design.

Is an Excel Bubble chart a 3-D scatter plot?

No. A Bubble chart places points using two axes and represents another value through circle size. A 3-D-looking chart option does not give the bubbles a genuine third data axis.

Related

Try it: the Date Difference Calculator: days, months, years and weekdays between two dates, with the DATEDIF formulas.

More in How-to basics

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.