Excel Chart Error Bars: How to Add and Format Them

Charts & pivot tables · Updated November 19, 2026

To add error bars in Excel for Microsoft 365 on Windows, select a chart, then choose Chart Design > Add Chart Element > Error Bars and pick an option. For values you have calculated yourself, choose More Error Bars Options, then Custom > Specify Value to select the positive and negative error ranges.

Add error bars to a chart

Error bars are visual markers that extend from chart points to show a range or amount of variability. They can make it easier to see whether measurements are precise, how much results vary, or how far a value may reasonably differ from its plotted estimate. They do not change the source data or the plotted point itself.

In Excel desktop for Windows, the basic process is:

  1. Click the chart, then click the data series that should receive error bars. If the chart has several series, select the intended series rather than assuming the whole chart is selected.
  2. Open Chart Design > Add Chart Element > Error Bars.
  3. Choose an available preset, such as Standard Error, Percentage, or Standard Deviation, or choose More Error Bars Options for additional controls.
  4. In the Format Error Bars pane, adjust the direction, end style, and error amount as needed.

Excel may add error bars to the selected series immediately when you choose a preset. If the result is not what you need, reopen the formatting pane and change the settings. On Excel for Mac, select the chart and use the Chart Design tab; its chart-element controls are similar, although ribbon placement and labels can vary by release. Some advanced chart-formatting controls may not be available in Excel for the web.

What the different error bars mean

The preset names describe how Excel calculates the length of an error bar, not what your study or experiment means. The right choice depends on the data and the question you want the chart to answer.

  • Standard Error: Excel calculates an error amount from the plotted series using a standard-error approach. Standard error is related to the spread of observations and the number of observations; it is often used to describe the precision of an estimated mean. It is not automatically a confidence interval.
  • Standard Deviation: This shows a standard-deviation-based amount for the series. Standard deviation describes variation among values, not directly the precision of the mean. Where Excel offers a multiplier, a value of 1 represents one standard deviation and a larger value extends the bars accordingly.
  • Percentage: Excel sets the error amount as a percentage of each plotted value. For instance, a 5% setting produces a different absolute-length bar for a value of 20 than for a value of 100.
  • Fixed Value: A single amount is used for the selected series. This can be useful when the measurement process has a known constant tolerance, provided the same amount is appropriate for every point.
  • Custom: You supply the error amounts, which is useful when each point has its own uncertainty or when the upper and lower ranges differ.

For example, if a column represents an average of repeated measurements, standard deviation can communicate how dispersed those measurements were, while standard error can communicate the estimated precision of the average under appropriate assumptions. Those tell different stories. Label the chart or its notes so readers know which statistic is shown and how it was calculated.

How to choose the right type

Start with the meaning of the data, not the appearance of the chart. Ask what the plotted value represents, what variability you measured, and what you want the reader to infer. If values are averages from repeated observations, decide whether the intended message concerns variation among observations, uncertainty in the estimated average, a confidence interval, or a measurement tolerance. These are not interchangeable.

Excel’s built-in presets are convenient but may not match a particular analysis. For example, a confidence interval generally requires a defined confidence level and an appropriate calculation based on the sample size and data assumptions. Selecting Standard Error does not instruct Excel to calculate a 95% confidence interval. If your analysis calls for a particular interval, calculate its positive and negative distances from each plotted estimate and use custom error bars.

Consider these practical cases:

  • For repeated measurements where the reader needs to see the spread of individual observations, consider standard deviation or show the individual points as well.
  • For means where the goal is to show uncertainty in those estimates, use the statistic specified by the analysis, such as a calculated confidence interval, and enter it as custom values when needed.
  • For an instrument with a documented plus-or-minus tolerance that applies equally to every reading, a fixed value may be suitable.
  • For values whose uncertainty is proportional to their size, a percentage may be a reasonable display choice if that proportional relationship is justified.

Do not choose a preset just because it makes the bars look balanced. State in a caption, axis note, or nearby text what the bars represent, including units and, where relevant, sample size and method.

Enter custom positive and negative values

Custom error bars let you plot unequal amounts above and below each point. This is important when uncertainty is asymmetric, when values have different sample sizes, or when you have already calculated confidence limits. First make two worksheet ranges: one for positive error amounts and one for negative error amounts. Each range should correspond point by point to the selected series. Enter distances, not the upper and lower endpoint values themselves.

For example, suppose a plotted estimate is 12 and its interval runs from 10 to 15. The positive error amount is 3 and the negative error amount is 2. Entering 15 and 10 as the error amounts would make the bars much too long, because Excel adds those amounts to the point rather than treating them as endpoints.

  1. Select the chart series and open Chart Design > Add Chart Element > Error Bars > More Error Bars Options.
  2. In the Format Error Bars pane, choose Custom under Error Amount, then select Specify Value.
  3. For Positive Error Value and Negative Error Value, select the corresponding worksheet cells, or enter the ranges in the boxes.
  4. Confirm the dialog and check that the first error amount aligns with the first plotted point, the second with the second point, and so on.

The positive and negative ranges need the same number of values as the plotted points in that series. A mismatch, a selection that includes a header, or ranges in the wrong order can produce an error or misleading bars. Custom amounts should be nonnegative distances; the direction is determined by the positive or negative range you assign.

Format direction, caps, and appearance

In the Format Error Bars pane, Direction controls whether bars extend in both directions, in the positive direction, or in the negative direction. End Style controls whether the ends have small caps. The available choices can depend on the chart type and the selected error-bar dimension. Use the pane’s Fill & Line controls to change color, line width, or dash style, and keep the bars visible without letting them overwhelm the data.

For a clean chart, use a contrasting but restrained line color, avoid excessive width, and check that bars do not disappear behind columns or markers. If a series uses many points, light lines may make the display easier to read. Use consistent styling when two series use the same statistical definition, but do not style unlike measures identically without labeling them clearly.

Excel can display horizontal and vertical error bars for supported chart types. In a scatter chart, horizontal bars can represent uncertainty in X values and vertical bars uncertainty in Y values. In many column or line charts, the usual error bars are vertical because the plotted uncertainty concerns the value axis. If the expected direction is unavailable, check the chart type and which series or axis is selected.

Chart types and series selection

Error bars are most useful on charts that show individual data points or summarized values, such as column, line, and XY scatter charts. The exact options depend on chart type. A chart may allow only one direction of error bars, or its layout may make a particular type difficult to interpret. If Excel does not offer the option you expect, try a supported two-dimensional chart rather than a three-dimensional chart, and inspect the selected series.

In a chart with multiple series, error bars are configured per series. You can give each series its own custom ranges and, where appropriate, its own error-bar definition. Be cautious when series have different units, sample sizes, or statistical meaning: comparing the visual lengths of their bars may not be meaningful. If your chart has secondary axes, confirm that the errors correspond to the intended series and axis.

For a scatter chart, verify whether the X and Y axes have the same units or scales before comparing horizontal and vertical bar lengths. A long-looking bar may reflect the axis scale rather than greater uncertainty in a meaningful common unit. Axis labels and units help prevent that misreading.

Common problems and how to fix them

  • Error bars appear on the wrong series: Click a data point or series to select it before adding the bars. In a chart with multiple series, confirm the selected series name in the formatting pane.
  • Custom bars are identical despite different data: Check whether a preset such as Fixed Value or Percentage is still selected. Choose Custom and assign both positive and negative ranges.
  • Bars are far too large: Check that custom cells contain distances from the point, not interval endpoints, and confirm the units match the plotted values.
  • Only one side appears: Inspect Direction and the positive and negative custom ranges. A one-sided setting may have been selected intentionally or accidentally.
  • Some custom bars look associated with the wrong points: Recheck the order and number of cells. Excel follows the order of the plotted series, which may differ from the visible worksheet order if the chart data were rearranged.
  • Error bars are hard to see: Change their line color or width, add caps, or adjust the chart scale. Avoid changing the scale in a way that exaggerates differences without making that choice clear.

If the pane is missing, select an error bar itself or use the chart-element menu to add one, then reopen More Error Bars Options. Clicking the chart background can select the chart area rather than a series, so selection is often the first thing to correct.

How to interpret them without overclaiming

Error bars are a visual summary, not proof that two groups differ or do not differ. Whether overlap between bars has a statistical interpretation depends on what the bars represent, how they were calculated, and the study design. Overlapping standard-deviation bars, standard-error bars, and confidence intervals have different implications. Do not use a quick visual comparison as a substitute for an appropriate statistical test when the question requires one.

Include enough context for a reader to interpret the display: identify whether the bars show standard deviation, standard error, a confidence interval, a range, or measurement tolerance; state the confidence level if applicable; and provide the sample size when relevant. Also consider showing the underlying observations. A mean with error bars can conceal skewed data, outliers, or very different distributions that individual points or a box-and-whisker chart would reveal.

Frequently Asked Questions

How do I add error bars to a bar chart in Excel?
Select the chart series, then choose Chart Design > Add Chart Element > Error Bars. Pick a preset or open More Error Bars Options to set custom amounts and formatting.

Can I use different error bars for each data point?
Yes. Choose Custom > Specify Value and select a worksheet range for positive amounts and another for negative amounts. Each range must align with the plotted points in the selected series.

Are standard error bars the same as 95% confidence intervals?
No. Standard error is not itself a 95% confidence interval. A confidence interval uses a specified confidence level and an appropriate calculation; enter its calculated distances as custom error bars if needed.

Why do my Excel error bars all have the same length?
You may have selected Fixed Value, or your custom ranges may contain repeated values. Check Error Amount in the Format Error Bars pane and confirm that the intended point-by-point ranges are selected.

Can error bars go both horizontally and vertically?
Some chart types, including XY scatter charts, can show both X and Y error bars. Other chart types may support only the direction appropriate to their axes. Check the options available for the selected chart and series.

Should error bars show standard deviation or standard error?
Choose based on the message: standard deviation describes spread among observations, while standard error concerns the precision of an estimated mean. Neither is universally correct, and the chart should label the chosen measure.

Related

Try it: the Percentage Formula Generator: percent of, percent change, increase and decrease, with the formula.

More in Charts & pivot tables

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

Spotted an error in this guide? Let us know.