LookML Dashboard Parameter Reference

This reference summarizes the key parameters used to build high-quality dashboards, as identified in the Looker demo environments.

Dashboard-Level Parameters

Parameter	Description	Recommended Usage
layout: newspaper	24-column grid layout.	Mandatory for professional dashboards.
preferred_viewer	Viewer engine.	Use dashboards-next for the modern experience.
crossfilter_enabled	Enables click-to-filter.	Set to true for interactive analytics.
auto_run	Execution behavior.	Set to false if the dashboard has >15 heavy tiles.
filters_bar_collapsed	Initial filter state.	Set to true to focus on data first.
filters_location_top	Filter bar placement.	Set to false to move filters to the right sidebar.
query_timezone	Timezone for queries.	Always use user_timezone.
tabs	List of dashboard tabs.	Use to group elements by persona or topic.

Element-Level Parameters

Parameter	Description	Best Practice
type	Visualization type.	Match type to the question (e.g., looker_area for trends).
width / height	Tile dimensions.	Use standard sizes (KPIs: 4x4, Trends: 12x8).
row / col	Tile position.	Maintain clear horizontal and vertical alignment.
tab_name	Tab assignment.	Must match a name defined in the top-level tabs.
listen	Filter mapping.	Ensure every tile "listens" to relevant dashboard filters.
note	Inline documentation.	Use note_display: hover for definitions.
row_total	Column-wise totals.	Use right to add a total column to grids.
show_totals	Display aggregate totals.	Essential for summary tables.
reference_lines	Static target lines.	Essential for providing context on trend charts.
fill_fields	Data gap filling.	Use for date dimensions to ensure trends have no breaks.

Filter Configuration (ui_config)

Type	Display	Best Use Case
button_group	inline	Dimensions with 2-5 options (e.g., Gender).
radio_buttons	popover	Mutually exclusive options.
dropdown_menu	popover	Dimensions with many options (e.g., Brand).
relative_timeframes	inline	The primary date filter for the dashboard.
advanced	popover	For complex multi-select or string matching.

Visual Styling (y_axes)

• Dual Axes: Use two objects in the y_axes list with orientation: left and orientation: right.
• Value Format: Use valueFormat: "$#,##0" or similar for professional labels.
• Log Scales: Use type: log for metrics with extreme variance (e.g., population data).