Pivot has a couple of key concepts to understand to get the most out of the tool.

Data cubes

Data cubes are exploratory views that map to datasources in Druid. One or more data cubes may be created from an indexed datasource in Druid.

Data cubes are focused around a primary visualization and are best used for rapid drill downs into data. By filtering or splitting on dimensions, you can generate a variety of different visualizations. You can additionally drill further into each visualization by filtering on any arbitrary combination of dimensions, or clicking around on the visualization itself. Each visualization may be added to a dashboard.


Filters narrow the field of view of data. Filters are applied on boolean expressions of dimension values, or on ranges of possible dimension values.


Splitting a single dimension shows measures for each of the dimension's values. Splitting on multiple dimensions produces a nested view of the data. For example, given two dimensions "country" and "city", splitting on "country" yields measures for individual countries such as USA, Canada, China, etc, and splitting on country and city yields measures for individual cities in each and every country.


Dimensions are attributes of data that you split or filter on.


Measures are aggregates of your data. You cannot split or filter on a measure. If you need to do so, you need to create a new dimension for the measure.


Dashboards contain multiple visualizations. Dashboards can be set to have a common filter across all visualizations.