Power BI technique · built-in features only
How to use native small multiples in Power BI
By Nawaf Sharaf · Founder & Reliability Engineer · Updated
When to use it. Use small multiples to compare the same shape across a handful of categories (sales trend per region). With more than about twelve panels, readers lose track; filter to the top ones.
Worked example
Worked example · Corvid Energy · MWh by month, one panel per site (shared scale)
Ashby
Jan · 410peak 415Dec · 415
Brent Moor
Jan · 388peak 392Dec · 392
Calder
Jan · 295peak 299Dec · 299
Dunmore
Jan · 240peak 280Dec · 243
Elsby
Jan · 182peak 184Dec · 184
Fenwick Lab
Jan · 21peak 38Dec · 21
Corvid Energy is a fictional operator with six sites. One line chart with Site in the Small multiples well becomes six panels of monthly MWh on one shared y-axis that tops out at 415 MWh. The four large sites follow the heating shape: high in winter, low in summer. Dunmore is the exception, peaking in July because its load is cooling.
Fenwick Lab also peaks in July, and its summer load is 81% above its winter low, but at this scale its line is nearly flat along the bottom of its panel. That is the shared scale doing its job (Fenwick is tiny next to Ashby) and also its cost: the shape you might care about disappears. When the question is "which sites have the same seasonal pattern", filter the small sites out or show them in a separate visual; when the question is "which sites matter", the flat line is the honest answer.
Steps
- 1
Start from a supported chart
Small multiples are available for bar, column, line and area charts (check Microsoft's page for the current list).
- 2
Add the split field
Drag the category (Region) into the Small multiples well in the Build pane.
- 3
Set the grid
Format → Small multiples → Layout: choose rows and columns so panels stay wide enough to read.
- 4
Format panel titles
Format → Small multiple title: position, font and size. Short category names work best.
- 5
Sort the panels
Use the visual's More options (…) → Sort axis to sort panels by the split field or a measure.
How it works
Small multiples split the visual's query by the field in the Small multiples well and draw one panel per value, all sharing the axes, legend and formatting. Because the panels share a scale, the eye can compare heights across panels directly, which is the whole point of the technique and the reason it beats six separate charts with automatic axes.
Panel order follows the visual's sort, so sort by a measure to put the biggest site first or by the split field for an alphabetical grid. The grid's rows and columns are set in the Layout options; Power BI scrolls when there are more panels than fit.
Small multiples work with the visual's other features: a legend splits each panel's series, and tooltips and cross-highlighting work per panel.
What Microsoft documents
Limits and behaviour from Microsoft Learn that change how you build this.
- Microsoft supports small multiples on bar, column, line and area charts. The grid starts at two by two and can go up to six by six; panels that don't fit load as the reader scrolls. Each row of panels shares one y-axis on its left and each column one x-axis at its bottom.
- Several features are switched off for now in a small-multiples visual: trend lines and forecasting in the Analytics pane, zoom sliders, axis zoom, total labels on stacked charts, concatenated axis labels, rectangle multi-select, and the Analyze and Summarize context menu items. Plan the page without them.
- A small-multiples chart also does not load more categories when the reader scrolls to the end of its category axis, so a long axis is cut rather than paged.
- Against a live connection, the source has to support second-generation DAX queries: Analysis Services 2016 or newer for tabular, 2019 or newer for multidimensional.
Variations
- For many categories (twenty sites or more), a heatmap-style matrix with conditional formatting compares more items in less space.
- Combine with a Top N filter on the split field to keep the grid to the biggest six or nine panels.
Check that it works
Count the panels against the number of values in the split field after filters, and sort the visual so the order is deliberate. Check that the one shared axis still shows something useful for the smallest panel; if it doesn't, decide whether that is the honest answer or whether small items belong in a separate chart.
Common mistakes
- Panels share an axis scale, which is what makes them comparable; a category with tiny values will look flat.
- Too many panels scroll; filter with a Top N filter on the split field.
- Legends repeat meaning across panels; keep series to two or three.
- A date hierarchy on the axis can repeat the year label in every panel; use a month column with Sort by column instead.
FAQ
Can small multiples be used with combo or scatter charts?
Check Microsoft's documentation for the current list of supported visuals; when a chart isn't supported, a custom visual or several aligned charts are the alternatives.
How do I show only the top regions?
Add a Top N filter on the split field in the Filters pane, by the measure in the chart.
Sources
Menu names follow Power BI Desktop as documented on Microsoft Learn; they can move between releases. The companies and figures in the worked example are invented.
Related
Assets and guides that go with this technique.
- Heatmap visual (an alternative for many categories)
- Sales dashboard templates
- Top N with Others recipe
More Power BI techniques
- Custom Conditional Formatting Icons
- Dynamic Titles
- Power BI Field Parameters: Switch Measures and Axes
- Calculation Groups
- Toggle Between a Chart and a Table with Bookmarks
- Drillthrough to a Detail Page
- Report Page Tooltips
- Sparklines
- Top N with 'Others'
- Sort Month Names Correctly
- Colour Bars and Cells by a Measure