A bar chart is a visual representation that uses rectangular bars, where the length of each bar represents each value.
You can use the bar chart visualization when you want to compare values over different categories or time periods. The visualization can display the bars horizontally or vertically, and can be customized to group or stack bars for more complex data analysis.
To create a bar chart visualization, you need a dataset containing one string or time field (or column) and at least one numeric field, though preferably more than one to make best use of the visualization.
The text or time field is used to label the bars or values in each row of data and the numeric fields are represented by proportionally sized bars.
If you have more than one text or time field, by default, the visualization uses the first one, but you can change this in the x-axis option as described in the [Bar chart options](#bar-chart-options) section.
If your dataset contains multiple rows, the visualization displays multiple bar chart groups where each group contains multiple bars representing all the numeric values for a row.
While the first field can be time-based and you can use a bar chart to plot time-series data, for large amounts of time-series data, we recommend that you use the [time series visualization](https://grafana.com/docs/grafana/latest/panels-visualizations/visualizations/time-series/) and configure it to be displayed as bars.
We recommend that you only use one dataset in a bar chart because using multiple datasets can result in unexpected behavior.
When the graph is vertically oriented, this setting rotates the labels under the bars. This setting is useful when bar chart labels are long and overlap.
The **Gradient mode** option located under the **Graph styles** has a mode named **Scheme**. When you enable **Scheme**, the bar receives a gradient color defined from the selected **Color scheme**.
##### From thresholds
If the **Color scheme** is set to **From thresholds (by value)** and **Gradient mode** is set to **Scheme**, then the bar color changes as they cross the defined thresholds.
{{<figuresrc="/static/img/docs/time-series-panel/gradient_mode_scheme_thresholds_bars.png"max-width="1200px"caption="Color scheme: From thresholds">}}
##### Gradient color schemes
The following image shows a bar chart with the **Green-Yellow-Red (by value)** color scheme option selected.
Enter a **Value** to change the size of the text on your bar chart.
## Axis
Use the following field settings to refine how your axes display.
Some field options will not affect the visualization until you click outside of the field option box you are editing or press Enter.
### Placement
Select the placement of the Y-axis.
#### Auto
Grafana automatically assigns Y-axis to the series. When there are two or more series with different units, then Grafana assigns the left axis to the first unit and right to the following units.
If you have more than one Y-axis, then you can give assign different labels with an override.
### Width
Set a fixed width of the axis. By default, Grafana dynamically calculates the width of an axis.
By setting the width of the axis, data whose axes types are different can share the same display proportions. This makes it easier to compare more than one graph’s worth of data because the axes are not shifted or stretched within visual proximity of each other.
### Soft min and soft max
Set a **Soft min** or **soft max** option for better control of Y-axis limits. By default, Grafana sets the range for the Y-axis automatically based on the dataset.
**Soft min** and **soft max** settings can prevent blips from turning into mountains when the data is mostly flat, and hard min or max derived from standard min and max field options can prevent intermittent spikes from flattening useful detail by clipping the spikes past a defined point.
You can set standard min/max options to define hard limits of the Y-axis. For more information, refer to [Standard options definitions](ref:standard-options-definitions).
