Filter, Sort and Aggregate Your Chart Data

Open a chart to land in the editor. The configuration panel sits on the right; if it's hidden, use the editor toggle in the chart header to open it. On a small screen, tap the same toggle to slide the panel in.

The panel is organized top to bottom: Chart type, the data section, Filters, Sort, and appearance settings. Every change saves on its own and the chart preview updates as you edit — there's no save button.

Note: Viewers can't edit charts. You need an Editor or Admin role on the workspace to change axes, filters, or sort.

In the data section, assign your dataset's columns to the chart:

• Axis X (Categories) — click Add dimension and pick the column for the horizontal axis (for example, "Month" or "Product"). If you pick a date column, a Date granularity dropdown appears so you can group by day, week, month, quarter, or year.
• Axis Y (Values) — click Add measure and pick the numeric column to plot. Add more than one measure to plot several series.
• Segment (Color by) — optionally click Add segment to split each category by a second dimension (for example, revenue by region). With a segment, only the first measure renders.

Each field shows a small type icon, and you can hover a field to preview sample values. Click the × on any field to remove it.

Because a chart groups rows by category, each measure needs an aggregation. Use the dropdown under the measure to choose one:

• Sum, Average, Min, Max — math over the numeric values in each group
• Count — the number of rows in each group
• Distinct — the number of unique values
• None — no grouping math; uses a single value per group

Numeric measures default to Sum. Counts work on any column, since Count tallies rows regardless of the column you picked.

In the Filters section, keep Visual selected on the Visual / SQL switch, then click Add filter and choose a column. Each filter row has:

1. The column (with a × to remove it).
2. An Operator — the choices depend on the column type. Text offers Contains, Equals, Starts with, Is empty, and more; numbers offer Greater than, Less than, Equals, and so on; dates add presets like Last 7 days and This month.
3. A value field, shown only when the operator needs one (operators like Is empty take no value).

Add several filters to stack conditions. Once you have two or more, an AND / OR switch appears to control how they combine. For advanced needs, flip the switch to SQL and write a raw WHERE clause instead.

By default categories keep their natural order from the dataset. To override that, use the Sort section: click Add sort field, pick a column, then choose Ascending or Descending. Sorting by a measure (for example, total revenue descending) turns a chart into a ranked view.

The Sort section appears for aggregated chart types and tables. It's hidden when a Segment is active, since segmented charts keep their category order automatically.

Filtering, sorting, and aggregating never change your data. ChartBrick stores each dataset as a read-only snapshot and runs your settings as a query against that snapshot at render time — your spreadsheet, database, or file is only ever read, never written.

That means you can experiment freely: try a different aggregation, stack filters, flip the sort, and nothing about the underlying source changes. To pull in new rows from the source, refresh the dataset instead.