Data Label: Meaning in a Graph

A data label is text printed directly on or next to a bar, point or pie slice that states its exact value — for example "42" above a bar or "35%" on a slice. In a graph it means the reader no longer has to estimate the number from the axis and gridlines; the value is written right there.

Charts are good at showing shape — which bar is tallest, which way a line is heading — but poor at communicating precise numbers. Data labels close that gap. Used well, they let one chart do the job of both a picture and a table. Used everywhere, they bury the picture in text.

Data vs data label: what is the difference?

The data is the underlying number in your table. The data label is that number (or something derived from it) displayed as text on the chart. The bar's length and the slice's angle are also drawn from the data; the label is simply a second, exact way of showing the same value.

On a pie chart this distinction matters because a label can show different things for the same slice. If a survey has 70 "Yes" answers out of 200, the slice can be labelled with:

• the value — "70";
• the percentage of the whole — "35%" (calculated: 70 ÷ 200);
• the category name — "Yes"; or
• a combination — "Yes: 35%".

All four are data labels; only the first is the raw data. For pie charts, percentages are usually the most helpful because the whole point of a pie is to show parts of a whole.

Data label vs other chart labels

When to use data labels

• Exact values matter. A report reader needs "42" rather than "a bit over 40".
• There are few data points. Up to roughly a dozen bars or a handful of pie slices can carry labels comfortably.
• Differences are small. If two bars are 48 and 51, labels make the gap explicit where the eye cannot.
• You want to drop the axis. With every bar labelled, the value axis and gridlines become redundant and can be removed for a cleaner chart.

Skip them, or label selectively, on dense line charts and scatter plots. Labelling every point on a 365-day line creates an unreadable smear; label just the latest value, the peak, or the point you are talking about instead.

How to place and format data labels

1. Position consistently. On bar charts, put labels just outside the end of each bar (or inside the end if the bars are long and the colour contrast is good). On pie charts, put labels inside large slices and outside small ones with a short leader line.
2. Round sensibly. Show only the precision that matters: "3.2M" instead of "3,214,567.00".
3. Include units where needed. "35%" or "$1.2k" avoids ambiguity, especially when the axis has been removed.
4. Keep contrast high. Dark text on light bars, light text on dark bars; never let a label sit half on and half off a shape.
5. Avoid overlaps. If labels collide, reduce the number of categories, rotate to a horizontal bar chart, or label only key values.

How to add data labels in common tools

Most chart tools have a single switch for this:

• Excel: select the chart, click the Chart Elements (+) button and tick Data Labels; the arrow beside it offers positions such as Outside End or Inside End.
• Google Sheets: open the chart editor, go to Customize → Series and tick Data labels.
• Chart.biz: the bar chart maker has a "Show values on bars" option that labels every bar, and the pie chart maker lets you label slices by percentage, value or name. Both run in the browser with no sign-up.

Related terms

Each data label describes a single data point. For a broader walkthrough of labelling, see how to label chart axes, and for choosing the right chart in the first place, how to choose a chart.