This entry is part of the Chart Index, the reference library for the Chart Design Field Guide.
A radial bar chart draws each series as an arc on its own concentric ring, sweeping clockwise from a common start angle by an amount proportional to its value. It is the chart of fitness trackers and progress dashboards — three rings closing towards completion — and it borrows the gauge's metaphor of a dial filling up while stacking several dials inside one another.
The form is compact and the metaphor is strong, which is why it persists. Its geometry is also quietly dishonest: the same value drawn on an outer ring covers more distance than on an inner ring, so the eye, which reads arc length, is shown a different chart from the one the angles encode. This entry sets out when the metaphor is worth that cost, and how to limit the damage when it is.
What it is
Each series occupies a ring. The ring's angular sweep encodes the value against a fixed maximum at the end angle; a light track behind each arc shows the full sweep, so the unfilled remainder is visible. Several categories within one series stack along the ring as differently coloured segments. The rings are named on a radial axis at the start angle; the value ticks run around the outside.
The hero chart above tracks six states against their annual sales target, three quarters in. Each ring is a state, each segment a quarter's contribution, and the full 270° sweep is 100 per cent of the year's target. Queensland has reached 88 per cent and will finish early; Tasmania is at 55 and will not. The tracks make the shortfall visible as a gap, which is the form at its best. Now compare the arc lengths: Tasmania's outermost ring at 55 per cent is drawn longer than New South Wales's innermost at 82. Randomise the values and watch which rings look big — it is the outer ones, whatever the data says.
When to use it
Radial bar charts are the right choice when:
- A small set of progress measures (two to six) shares a fixed and meaningful maximum — a target, a capacity, 100 per cent.
- The completion metaphor is the message: how much of the ring is filled, how much remains.
- The display is compact and the exact values are also printed or available on hover.
- Rings are ordered deliberately, so the reader is not asked to compare arcs at different radii.
When not to use it
- Values do not share a scale or maximum. Rings with different denominators cannot be read against each other at all.
- Readers need to rank similar values. Two rings at 62 and 68 per cent are indistinguishable once radius has had its say. Use bars or bullets.
- There are more than six rings. The inner rings shrink to stubs and the outer rings dominate.
- The chart needs a legend, many tracks and many colours to decode. That is a table wearing a costume.
Design principles
Know the radial-length distortion
The pair of figures below is the article's central argument. Four states, four values, and the smallest value is on the outermost ring. On the radial bar, that ring draws the longest arc; the eye ranks the states in almost exactly the wrong order. The bar chart beneath it shows the same four numbers on a common baseline, where length and value agree.
Show the full track
When arcs represent progress, display the remaining track quietly so the denominator is visible. Without the track the reader sees an arc of some length; with it they see a ring that is part full, which is the only reading the form does well.
Use a fixed scale
Radial progress charts are misleading when the sweep is scaled to the largest ring rather than to the target. With an automatic maximum the best-performing series always closes its ring and looks complete, whatever it actually achieved. State the common range and pin it.
Order the rings and leave a gap
Because radius biases the reading, put the series the reader should compare first on adjacent rings, and put the largest value on the outside so the length bias and the data agree rather than fight. Leave the sweep short of a full circle — 270° is the house default — so the radial axis has somewhere to name the rings and a ring at 100 per cent is visibly distinct from one at 0.
Label the value directly
Place the value near its arc or in a clearly associated list. Arc length alone should not carry a precise business measure; hover for the exact figure, and print the number in the caption or beside the ring when the chart stands alone. The bullet chart below carries the same six values with their targets, in less height, with every value on a common straight baseline — it is the form to reach for once the completion metaphor has done its work.
Avoid decorative needles and bevels
Do not turn a radial bar into a simulated instrument panel. Flat arcs, a quiet track and one accent colour preserve the data and remove visual noise; gradients, drop shadows and needle hubs add chrome to a form that is already spending most of its pixels on geometry.
Anatomy
Tracks provide the scale, arcs encode values through angle, concentric rings separate series, and the start and end angles define the available sweep. The radial axis at the start angle names the rings; the circular axis at the outside carries the value ticks.
Related types
- Polar bar chart — the other way to bend a bar: categories arranged by angle, values by radius.
- Polar range bar chart — angular categories whose bars encode lower and upper radial bounds.
- Gauge — a single radial indicator; the radial bar is several gauges nested.
- Bullet chart — compact actual-versus-target comparison on a straight baseline; the precise alternative.
- Bar chart — the default whenever the completion metaphor is not needed.
Reading list
- Few, S. (2006). Information Dashboard Design. On compact, comparable performance displays and the cost of dial metaphors.
- Cleveland, W. S. & McGill, R. (1984). Graphical Perception: Theory, Experimentation, and Application to the Development of Graphical Methods. The ranking that puts angle below length.
- Munzner, T. (2014). Visualization Analysis and Design. On position and length encodings in radial layouts.
- Wilke, C. (2019). Fundamentals of Data Visualization. On visualising proportions and when a circular layout earns its place.