A gauge chart is a visualization that displays a single value on a dial or arc against colored ranges, showing at a glance whether a metric is below, within, or above its target zone. It borrows its metaphor from a car’s speedometer: a needle (or filled arc) pointing at the current value, with the scale behind it divided into bands — typically something like red, yellow, and green.
When to use a gauge chart
Gauges do one thing well: they answer “is this number okay?” without making the viewer think. That makes them a fit when:
- There’s a meaningful target or threshold. An NPS score, a service-level target, a utilization ceiling. The colored ranges encode the judgment — anything in the green band is fine — so viewers don’t need to remember what “good” looks like.
- The audience checks status, not detail. On a wall-mounted dashboard or an executive summary, a gauge communicates state in about a second.
- The value is bounded. Percentages, scores on a fixed scale, capacity used — gauges work best when the scale has natural ends.
When not to use one
The gauge’s defining weakness is that it hides the trend. A dial showing 87% tells you nothing about whether that’s up from 60% last month or down from 99% — and for most decisions, the direction matters more than the snapshot. If the question is “how is this metric doing?”, a line chart with a goal line, or a KPI number with a trend comparison, carries strictly more information in similar space.
Gauges are also expensive per pixel. One number occupies the area where a chart could show twelve months of history. A dashboard tiled with six gauges usually serves its readers worse than six compact trend cards would. Finally, resist inventing thresholds just to fill in the color bands: if nobody has actually agreed on what “red” means for a metric, the gauge is decorating a number, not adding judgment to it.
A reasonable rule: use a gauge when the threshold is the message, and a trend chart when the movement is.
Gauge charts in Metabase
Metabase has a native gauge chart type. Point it at a question that returns a single number — a count, a sum, or a saved metric — and define the ranges in the visualization settings, choosing the min, max, and the colored bands in between. You can label each band so the chart says “Needs attention / On track / Ahead” instead of leaving colors to interpretation.
On a dashboard, gauges respond to filters like any other card, so one gauge can report the conversion rate for whichever region or period a viewer selects, and drill-through lets them get from the dial to the underlying records. See the gauge docs for configuration details.
Related terms
Further reading
Put it to work
- NPS — Metric
- Conversion rate — Metric
- Sales pipeline dashboard — Dashboard
