Small Multiples

Small multiples are a grid of small charts that all share the same chart type, axes, and scales, with each panel showing one subset of the data. Because every panel is drawn the same way, the reader compares groups simply by glancing across the grid — the way you might scan a row of photographs. They are also known as trellis charts or faceted charts, and they are one of the most reliable ways to show many groups at once without the chaos of stacking everything onto a single plot.

What small multiples are

Take one chart — say a line chart of a value over time — and instead of plotting every group as a separate line on top of each other, draw a separate copy of that chart for each group and arrange the copies in a grid. Each small chart, or panel, contains the data for exactly one group. Every panel uses the same axes, the same scale, and the same size, so the only thing that differs from panel to panel is the data itself.

That sameness is the whole point. Once the structure is identical everywhere, your eye stops decoding axes and starts comparing shapes. A panel that rises steeply stands out next to flat ones; a panel with a dip is obvious. The reader does the comparison almost without effort.

Why they work

The alternative — cramming many series onto one chart — runs into a hard limit fast. Three or four lines can usually be told apart by color. Beyond that, lines cross and overlap, the legend becomes a memory test, and the chart turns into what is often called a "spaghetti chart." Small multiples sidestep the problem entirely: nothing overlaps, because each group lives in its own panel.

They also lean on a strength of human vision. People are good at noticing when one item in an otherwise uniform set is different. When every panel is the same size and scale, an unusual one — the steep climb, the lone decline — pops out the moment you scan the grid.

A worked example

Suppose you are tracking the same metric over six months for four groups. As one chart, that is four overlapping lines and a legend to memorise. As small multiples, it is four tidy panels on a shared scale — and the differences between groups read instantly. The grid below shows the idea: each panel has the same baseline and the same height, so the slopes are directly comparable.

When to use them — and when not to

Reach for small multiples when you have several groups and the comparison across groups is the point. They suit a handful up to a few dozen panels; past that the grid gets too small to read and a different summary may serve better.

• Good fit: the same measurement across regions, products, teams, or categories; one chart per group with a shared scale.
• Less good: only two or three groups — a single chart with a few lines or grouped bars is simpler and lets the reader compare values directly.
• Poor fit: the reader's main task is reading exact values rather than comparing shapes — a table may be better.

If you are unsure whether the answer is one chart or many panels, our how to choose a chart guide works through the decision starting from what you want the reader to see.

Getting small multiples right

• Use one shared scale. This is non-negotiable — different scales per panel break the comparison and can mislead badly.
• Keep panels identical. Same type, size, axes, and colors. Let only the data vary.
• Label each panel clearly with its group, and label the shared axes once where it is obvious they apply to all.
• Keep each panel simple. Small panels have no room for clutter, so light gridlines and minimal decoration matter even more here.
• Sort the panels meaningfully so their arrangement adds information rather than noise.

The most common form uses a line chart per panel, but the same idea works with bars, scatter plots, or area charts — anything where comparing the same chart across groups tells the story. Try it with your own data using the free, in-browser makers at make a chart.