How to Read an Excel Box and Whisker Plot Step by Step

To read an Excel box and whisker plot, match each box to its category, read the median line and the box edges against the value axis, then check the whiskers and any separate outlier dots. This reveals the middle spread, typical value, and unusually high or low observations in each group. The guide below applies to Excel for Microsoft 365 on Windows.

What You’ll Need

Excel for Microsoft 365 on Windows, or a recent Excel version that supports Box and Whisker charts.
A worksheet with numeric observations. Put each group in its own column and add a group name in the first row.
Enough observations in each group to make a distribution worth comparing. The chart can be created with fewer values, but its summary is more informative with a useful sample.

You do not need to calculate quartiles before creating the chart. Excel calculates and displays the summary from the data you select. If you already have a chart, you can start at Step 3.

Step-by-Step

Check how your data is arranged. Use one column per group, with the group name at the top and one numeric observation in each row below it. For example, columns might be labelled Store A and Store B, with individual transaction times beneath each heading. Each number should represent one observation, not a subtotal or average. Keep the measurements in the same units across groups. A box plot compares distributions, so mixing seconds with minutes or dollars with cents would make the visual comparison misleading. You can leave blank cells where a group has fewer observations, but do not fill gaps with zero unless zero is a genuine measurement.
Select the data and insert the chart. Include the group headings and the numeric observations, but not unrelated totals or notes. On the Windows ribbon, choose Insert > Insert Statistic Chart > Box and Whisker. If that exact button is not visible, look for the Statistical Chart menu in the Insert tab. Excel creates a chart with one box for each data series. If the categories appear in the wrong direction, select the chart and choose Chart Design > Switch Row/Column. For the recommended layout, each column should be one group.
Identify the value axis and the groups. Read the category labels along the horizontal axis to see which box belongs to which group. The vertical axis shows the measurement scale. Locate a box’s position on that scale before interpreting its height: a box higher up represents larger values, while a taller box represents greater spread in the middle of the data. Those are different observations. If the chart has no useful axis title, add one through Chart Design > Add Chart Element > Axis Titles, then label the measurement and units.
Find the median line. The horizontal line inside a box is the median, or 50th percentile. About half of the observations are at or below it and about half are at or above it. It is a measure of the distribution’s centre, but it is not necessarily the arithmetic average. For example, if one group’s median is 18 minutes and another’s is 24 minutes, a typical middle observation in the first group is lower. The median alone does not tell you whether the groups have similar variation or whether they contain outliers.
Read the lower and upper edges of the box. The bottom edge is the first quartile, Q1, and the top edge is the third quartile, Q3. Roughly one quarter of the observations fall below Q1, half lie between Q1 and Q3, and one quarter lie above Q3. The box therefore contains the middle 50 percent of the data. Read both edges against the value axis; do not assume the box begins at zero. A box stretching from 12 to 20 has an interquartile range of 8 units, because the interquartile range, or IQR, is Q3 minus Q1.
Interpret the box’s height and the median’s position. A short box means the middle half of the observations are relatively close together; a tall box means they are more spread out. If the median line is near the middle of the box, the lower and upper halves of that middle range have similar widths. If it sits closer to one edge, the two halves differ in spread. This can suggest asymmetry, but it does not by itself prove a particular distribution shape. Consider the whiskers and individual points too.
Follow each whisker to its endpoint. The lines extending from the box are whiskers. In Excel’s standard box plot, they reach to the most extreme observed values that are not classified as outliers, using the 1.5-times-IQR rule. In other words, the upper whisker usually reaches the largest data value no higher than Q3 plus 1.5 times the IQR, and the lower whisker reaches the smallest value no lower than Q1 minus 1.5 times the IQR. The whisker endpoint is an actual observation, not automatically the theoretical boundary. If no observations fall beyond those limits, the whiskers reach the minimum and maximum values.
Check for outlier points. A separate dot beyond a whisker represents an observation outside the 1.5-times-IQR limits. It is flagged as unusual relative to the rest of that group; it is not automatically a mistake or a value that should be deleted. For example, suppose a chart’s Q1 is 10 and Q3 is 22. The IQR is 12, so the upper fence is 40. A value of 45 is beyond that fence and may appear as an outlier. Investigate the source and context before deciding what to do with it. A valid rare event can be important to keep.
Review Excel’s chart options before comparing quartiles. Right-click a box and choose Format Data Series to inspect the series settings. Excel offers an inclusive or exclusive median quartile calculation. These choices can produce different quartile positions, especially with small samples, so use the same setting when comparing charts and document the choice if the result matters. The same panel can show or hide inner points, outlier points, mean markers, and a mean line. A mean marker is an average and is different from the median line. If dots are missing, check whether outlier points or inner points have been turned off.
Compare groups in context. Look first at median positions, then at box heights and whisker lengths, and finally at outliers. A group with a higher median has a higher midpoint, but it may also have much more variation. Overlapping boxes do not establish that two groups are identical, and separated medians do not establish why they differ. Check sample sizes, collection methods, time periods, and units before drawing a conclusion. A box plot summarizes the observed values; it does not explain causes or show every detail of the distribution.

On Excel for Mac, the chart is also available from the Insert tab’s Statistical Chart options in supported versions, though the ribbon layout and wording can vary by release. Excel for the web may display an existing box plot, but chart creation or formatting options can be more limited than in the Windows desktop app. If you cannot find a quartile or point-display setting, open the workbook in desktop Excel.

Common Mistakes

Calling the box the full range. The box covers the middle 50 percent, not every observation. The whiskers and any outlier dots show information outside that middle range.
Assuming the whisker always ends at the minimum or maximum. With the standard 1.5-times-IQR rule, a more extreme observation is plotted separately as an outlier. Do not describe a whisker as the full range if there are dots beyond it.
Confusing the median and the mean. The line in the box is the median. A mean marker or mean line appears only if that option is enabled. The two measures can differ, particularly when a group has skewed values or extreme observations.
Treating every outlier as bad data. First check whether it is a typing error, a measurement problem, or a valid unusual observation. Removing a genuine value can conceal an important feature of the data.
Comparing charts with different settings or scales. Inclusive and exclusive quartiles can alter box edges, and different axis ranges can make similar groups look unlike one another. Match the quartile setting and inspect the axis labels before comparing.
Reading box height as sample size. A taller box indicates a wider middle 50 percent, not more observations. The plot’s box size does not communicate how many records were collected.

Tips

Keep the same units, measurement method, and time window for every group. Label units directly in the axis title so the scale is unambiguous.
Use clear category names and a descriptive chart title. If group names are long, widen the chart or simplify the labels rather than leaving readers to guess which box is which.
When comparing several groups, keep them on one shared value axis. Separate charts with different scales can exaggerate or hide differences.
For a closer look at a point, select it in the chart or hover over it to view the data-point tooltip when Excel provides one. If you need to identify the original record reliably, keep a copy of the source data and use the value and group to find the matching row.
Show the mean marker only when it helps answer the question. A mean can add useful context, but the median and mean should be labelled or explained so readers do not mistake one for the other.
Preserve the underlying observations alongside the chart. A box plot is a summary, so readers may need the data table to investigate sample size, repeated values, or the reason for an unusual point.

Troubleshooting

The chart shows one box instead of several: Check the data orientation. If each group is in a column, use Chart Design > Switch Row/Column if Excel interpreted rows as series. Confirm that the heading row is included and that the group data are arranged consistently.

The boxes or category labels look wrong: Check for stray text, totals, dates, or notes inside the selected range. Move summary rows outside the observations, and make sure each column contains one kind of measurement. A date is stored as a number in Excel, so an accidentally included date can distort the scale rather than look like ordinary text.

There are no visible outlier dots: There may genuinely be no values beyond the whisker limits, or the display option may be off. Right-click a box, choose Format Data Series, and turn on Show outlier points. Also check that points are not hidden by the chart size or overlapping markers.

The quartiles do not match another tool or a hand calculation: The tools may use different quartile conventions, and small samples can make the difference noticeable. In Format Data Series, compare the inclusive and exclusive median settings. Confirm that both calculations use the same observations and that blanks or invalid values have not been handled differently.

The chart looks flat or hides differences: Inspect the vertical axis bounds and units. A few very large values can compress most boxes near the bottom. You can adjust the axis bounds by right-clicking the value axis and choosing Format Axis, but do not crop the scale in a way that misleads readers. State clearly when an axis does not start at zero.

The Box and Whisker command is unavailable: Confirm that you are using an Excel version that supports this chart type and that you are in the desktop app. Some older editions and web interfaces may not expose the same chart options. Updating Excel or opening the workbook in Microsoft 365 for Windows may provide the full chart controls.

Frequently Asked Questions

What does the line inside an Excel box plot mean?
The line marks the median, or 50th percentile. Half the observations are at or below it and half are at or above it.

What do the whiskers represent in Excel?
They extend to the most extreme observed values within the standard 1.5-times-IQR limits. Values beyond those limits are treated as outliers and can appear as separate points.

Does a box and whisker plot show the average?
Not by default as the main line. The line is the median. Turn on the mean marker or mean line in Format Data Series if you also want to show the arithmetic average.

Why does Excel show different quartiles from my calculation?
Excel lets you choose inclusive or exclusive median quartile calculation. Other tools may use a different convention, especially for small data sets. Check the setting and compare the same values.

Can a box plot show the number of observations?
Not clearly from the box itself. A standard box plot summarizes quartiles and potential outliers, but the box’s height does not represent sample size. Include the sample counts in labels or a nearby table if readers need them.

Should I remove outliers before making a box plot?
Not automatically. Verify suspicious values against the source, then decide based on the purpose and data rules. Keep valid observations and explain any exclusions so the chart remains transparent.

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.