Skip to content
Saiku Cloud

The Saiku workbench

The Saiku workbench is the screen that opens when you click Open in Saiku on a cube from the Analyze page. It’s a four-region layout: cubes and fields down the left, drop zones across the top, a query toolbar above the result, and the result table (or chart) underneath. This page walks the regions so the rest of the Saiku UI docs make sense.

The left sidebar

The leftmost column shows your cubes and, once you’ve picked one, its fields.

Cube list. When the workbench opens you see every cube available to your account. Click a cube to load its fields below; the workbench locks onto that cube until you open another one.

Dimensions. Each dimension folds open to reveal its hierarchies, and each hierarchy opens to reveal its levels. A typical date dimension looks like Time › Year › Quarter › Month › Day — five levels, but you only use the ones you want.

Measures. Listed at the bottom. These are the numbers — sales totals, counts, averages — that the cube knows how to aggregate along whichever dimensions you drag in.

The drop zones

Across the top of the result area:

Columns. What you want side-by-side in the result. Drag a measure here to get one column per measure; drag a dimension level to get one column per level value.

Rows. Same idea, vertical. Most analyses put the dimensions you want to slice by here and the measures across.

Filters. Dimensions you want to restrict the query by without showing them in the output. “Only show 2024 data” goes here — drag the year level into Filters, then pick the year(s).

You drag fields from the sidebar into a drop zone. To remove a field, drag it off the drop zone, or right-click for a context menu.

The query toolbar

The bar above the result table holds the controls you’ll reach for most often during analysis.

  • Run query — triggers the query manually. Off by default (Auto-run is on); useful when you want to compose a complex pivot before committing.
  • Toggle auto-execution — flip auto-run on or off. With auto-run on, every drag-and-drop fires a fresh query.
  • Toggle fields — show or hide the drop zones at the top. Hide them once you’re done composing to maximise the result area.
  • Toggle sidebar — same for the left sidebar.
  • Swap axes — exchange Rows and Columns in one click.
  • Non-empty — hide rows or columns with all-null measures. Usually you want this on.
  • Show MDX — open a panel with the MDX query Saiku generated. Useful for debugging or for power users who want to hand-edit (see the MDX & export page).
  • Group parents — flatten parent levels when drilling, so the parent label doesn’t repeat for each child.

The result area

The bottom region renders the query output. Two views:

Table (default). A pivot grid with totals at the edges. Each cell shows one measure for one combination of dimensions. Click a parent value (e.g. “2024”) to drill down to its children.

Chart. Switch via the chart-type buttons on the toolbar. Bar, line, area, pie, sunburst, treemap, heatmap, waterfall, and several variants for showing multiple series at once. See Charts and visualisations for guidance on picking the right type.

Tabs

The bar across the top of the workbench manages multiple queries at once — each tab is an independent query against any cube. Open a second tab to run a comparison without losing your first analysis.

Save, open, share

Three buttons in the top-left handle workbook lifecycle:

  • New query — empty workbench on the current cube.
  • Open query — pick from saved workbooks; yours and ones shared by teammates.
  • Save query — commit the current state. Save into your private folder or a shared folder visible to the rest of your workspace.

Saved workbooks include the cube, the field placement, filters, sorts, the current chart view if any, and the workbook title. They’re identical to the original when you reopen them.

Where to go next