This page describes Looker’s new dashboard experience. For information on legacy dashboards, visit the Adding filters to legacy dashboards documentation page. To determine which type of dashboard you are using, compare the screenshots on the Viewing dashboards and Viewing legacy dashboards documentation pages.

This page is about standard dashboard filters; to learn about dashboard cross-filters, visit the Cross-filtering dashboards documentation page.

Understanding dashboard filters

Dashboard filters let a viewer narrow a dashboard’s results to only the data the viewer is interested in. Dashboard filters can apply to all tiles on a dashboard, or to only one dashboard tile.

When viewing a dashboard, a user can temporarily update dashboard filter values, which doesn’t affect any other users. This differs from filters that are part of a query underlying a dashboard tile; viewers can’t change these filters unless they have permission to edit the tile or to Explore from Here.

To make permanent changes to filters that all users will see, or to add or delete filters, you must have the Manage Access, Edit access level for the dashboard and be in edit mode.

Dashboard filters override filters applied to a tile’s underlying Look or Explore when both types of filters are filtering on the same field. See the Overriding Look filters in the Explore section with dashboard global filters Looker Community post for more details.

Dashboard filters and LookML

If a model, an Explore, or a field uses the label (model), label (Explore), or label (field) parameter, respectively, to change its name in the Explore menu or field picker, the name given in the label parameter is used in the Add Filter window.

If you are a LookML developer and are in Development Mode, you can create a dashboard filter on a field that appears only in your Development Mode version of a model, but the filter will not function as expected when you exit Development Mode or when the filter is used by other users. Similarly, if a filter field is deleted or renamed after the filter is created, the filter will not function as expected.

Adding dashboard filters

To get started, make sure the dashboard is in edit mode and that you have at least one query tile or Look-linked tile. Click Filters in the top toolbar, and then click Add Filter.

An Add Filter window appears next.

The Add Filter window

The Add Filter window appears pre-populated with fields (dimensions or measures) from any Explores that were used in the creation of this dashboard’s tiles.

For example, the following Add Filter window shows the Explores that are presently used in the dashboard’s tiles: the Flights Explore from the E Faa model and the Order Items Explore from the MB ecommerce_demo model.

Clicking the arrow to the left of an Explore expands the Explore to show the fields that are contained within it. To filter on a field from one of these Explores, you can use the Filter by dialog box to search for fields or manually enter a field name. You can also click a field name, such as Order ID, to add it to the Filter by dialog box:

If you select a field that is already being used by another filter, Looker will alert you. While you can still create the filter, you cannot apply multiple filters that use the same field to the same dashboard tile.

If you want to create a filter on a field from an Explore that is not used in the dashboard, see the Advanced filters section on this page.

If you have already selected a field in the Filter by dialog box of the Add Filter window, clicking the Advanced switch results in no changes to the Add Filter window and an advanced filter is not initiated.

Once you’ve selected the field you want to filter by, click Add.

Advanced filters

Advanced filters let you create dashboard filters using fields that are outside of the Explores used to create tiles on the dashboard. Click the Advanced switch to use advanced filters. This updates the Add Filter window with options to select a model, an Explore, and a field with which to create the filter:

First, select a model name from the drop-down Model menu. The drop-down menu contains all LookML models that you have access to.

Once a model is selected, the Explore dialog box populates with a drop-down menu of all the Explores that are present in the selected model.

Once the model and the Explore are selected, the fields that are available in that Explore will populate the Add Filter window. You can use the search bar to find the field that you want to filter by, or you can click a field name to select it.

Once you’ve selected the field you want to filter by, click Add.

Customizing filter configuration

Once you select a filter field and click Add, the Add Filter window shows a variety of customizable settings:

1. Title: Enter the title of the filter as you want it to appear on the dashboard. The title option pre-populates with the name of the filter-by field.
2. Control: Select from a list of control types, which vary depending on the type of data you are filtering. See the Dashboard filter controls section on this page to learn more about the different types of controls.
3. Display: For controls that can be displayed as either inline or popover, select how the filter will be displayed. See the Displaying dashboard filters section on this page to learn more about the different ways to display filters. If a control can be displayed only in one way, this option will not appear.
4. Values: To set specific value options for the filter, choose from the drop-down or enter the value options in this field. Leave blank to allow value options from the database to be surfaced, up to the maximum number of values available for that control. For numeric data, this field is replaced by the Min and Max fields.
5. Configure Default Value: Optionally, set the default value for the filter.
6. Require a filter value: Select the checkbox to require a value for the filter. See the Requiring a filter value section on this page to learn more about requiring filter values.
7. Select filters to update when this filter changes: Select the checkbox to link other filters to this filter. See the Setting up linked filters section on this page to learn more about linked filters. If there are no other filters on the dashboard, this option will be disabled.
8. Add and Cancel: Click one of these buttons to save or cancel the new filter.

The Tiles To Update tab lets you determine which tiles listen to the filter. Looker begins by automatically applying the filter to any tiles that were created from the same Explore as the filter and sets the value of Field to Filter to the same field as the field chosen for the filter:

1. Select All or None to turn the filter on or off for all tiles.
2. In the Field to Filter section, for each tile, choose which field will be affected by the filter, or choose not to apply the filter to an individual tile.
3. If you select a field that is already used in an existing filter, any tiles filtered by the existing filter are unavailable to the new filter.
4. Select Add or Cancel to save or cancel the new filter.

If any new tiles are added to the dashboard after the filter is created, if they were created from the same Explore as the filter, the dashboard filter is automatically applied to that tile.

If dashboard filters are applied to a merged results tile and the merge changes in any way — for example, if someone reorders the merge queries or adds or removes queries — all dashboard filters will be turned off for that tile. To reapply the filters, you must turn on the tile under the Tiles To Update tab of the Add/Edit Dashboard Filters window.

Dashboard filter controls

Dashboard filter controls let you customize the appearance of filters for dashboard viewers. The filter control types available in the Control drop-down as you create a dashboard filter depend on the LookML data type assigned to the field you’re filtering on.

To see the data type for a field you are filtering on, you must have the see_lookml permission. You can see the field’s LookML by selecting Go to LookML from the gear menu to the right of the field in the field picker.

Control	Description	Appearance	Data Types
Multiple Selection
Button Group	This type of filter is useful for giving viewers a curated set of limited options. Up to 30 options can be curated in the Values setting. For string, tier, and zipcode data types, if the Values setting is left blank, the first 30 values from the database are shown. A button is colored once it’s selected.		s n t zc yn di du
Checkboxes	Like a button group but with a slightly different design, this type of filter is useful for giving viewers a curated set of limited options. Up to 50 options can be curated in the Values setting. For string, tier, and zipcode data types, if the Values setting is left blank, the first 50 values from the database are shown.		s n t zc yn di du
Tag List	A combination of drop-down and checkbox filters. The drop-down options can be curated in the Values setting. For string, tier, and zipcode data types, all values can be surfaced from the database if the Values setting is left blank. Viewers can expand the drop-down by clicking on the chevron. Taglist controls can appear only in the popover orientation or behind the More button.		s n t zc di du
Range Slider	Range slider minimum and maximum can be set in the Settings tab of the filter configuration window. Viewers can set the filter value range by adjusting both ends of the slider.		n di du
Single Selection
Button Toggles	This type of filter is useful for giving viewers a curated set of limited options. Up to 30 options can be curated in the Values setting. For string, tier, and zipcode data types, if the Values setting is left blank, the first 30 values from the database are shown. A button is colored once it’s selected.		s n t zc yn di du p
Radio Buttons	Like a button toggle but with a slightly different design, this type of filter is useful for giving viewers a curated set of limited options. Up to 50 options can be curated in the Values setting. For string, tier, and zipcode data types, if the Values setting is left blank, the first 50 values from the database are shown.		s n t zc yn di du p
Drop-down Menu	The drop-down options can be curated in the Values setting. For string, tier, and zipcode data types, all values can be surfaced from the database if the Values setting is left blank. Viewers can expand the drop-down by clicking on the chevron and then either selecting an option from the drop-down or typing in the box to narrow the drop-down options. Viewers can also select Any value. Drop-down menu controls can appear only in the inline orientation or behind the More button.		s n t zc yn di du p
Slider	Slider minimum and maximum can be set in the Settings tab of the filter configuration window. Viewers can set the filter value by adjusting the right end of the slider.		n di du
Dates and Times
Single Day	Viewers can click the date to expand the calendar and select a new day. Single day menu controls can appear only in the inline orientation or behind the More button.		dt* The single day control can be used with most timeframes and time-based types, but viewers can only select single dates with this type of control.
Date Range	Viewers can click on the date range to expand the calendar and select a new date range. Date range menu controls can appear only in the inline orientation or behind the More button.		dt* The date range control can be used with most timeframes and time-based types, but viewers can only select date ranges with this type of control.
Timeframes	Viewers can either select from a set of timeframes built into Looker or create their own custom timeframes by clicking the Custom tab. See the Viewing dashboards documentation page to see the full range of timeframes options for a Timeframes filter. Timeframes options such as Last 7 Days include the current day. For a timeframe that excludes the current day, select Advanced from the control drop-down and configure the value using complete days. Year to Date includes the beginning of the year through to the current second. Viewers can click on the timeframe to see the timeframes options. Timeframes menu controls can appear only in the inline orientation or behind the More button.		dt* The timeframe control can be used with most timeframes and time-based types, but viewers cannot select time options with this type of control.
Other
Advanced	See the Using advanced controls section on this page to learn more about advanced controls. Advanced controls can appear only in the popover orientation or behind the More button.		s n t zc yn dt di du l p
*Note about date and time data: Some timeframes and time-based types are interpreted as different data types when Looker is selecting control options for them. Two examples are the yesno timeframe, which is interpreted as a yesno data type and therefore has the control options available to the yesno type; and the hour_of_day timeframe, which is interpreted as a number data type and therefore has the control options available to the number type.

Using advanced controls

An advanced control can be applied to all types of data and provides some additional flexibility in the filter conditions you can set up. To create an advanced control, make sure you have selected Advanced in the Control field of the filter configuration window.

The Configure Default Value field populates with appropriate filter condition options for the type of data you’re filtering. For example, the following filter is filtering on product category, which is a string data type, so the filter condition options include is, contains, starts with, and so on.

Filters on other types of data, such as number or date types, will display different condition options.

The Allow multiple filter values option allows users to select more than one value for the filter. It defaults to enabled. To restrict users so that they can select only one value, disable this option.

Once you select your filter condition option, you can select a value from the drop-down to the right, which is populated with values from your database. Or, you can type a value to receive a suggestion.

You can also enter a value yourself and click Enter to create a value.

You can click the + next to your filter to add new conditions to the filter, which will be added as either OR conditions or AND conditions, depending on the types of conditions and values. To remove a condition, click the X next to that condition.

Advanced controls, dates, and times

While there are several filter controls that allow filtering on dates, you can also use advanced controls with dates. Among other things, using advanced controls lets users select “complete” ranges that do not include the current time period, by selecting complete days, complete weeks, and so on, in the condition options. This is unlike the timeframe filter control, which includes the current time period.

Advanced controls also let users select time values, such as hours, minutes, seconds, and so on.

Matches a user attribute

Advanced controls also give you the option to select matches a user attribute. This lets you set filters that change dynamically based on dashboard viewers’ user attributes. See the Filtering and limiting data documentation page for more information.

Matches advanced

The final option in the Configure Default Value option drop-down is matches advanced. With this option selected, you can enter a Looker filter expression to customize a filter beyond the options provided in the options drop-down.

You can also reference user attributes in matches advanced filters by using the following syntax:

{{ _user_attributes['name_of_your_attribute'] }}

Displaying dashboard filters

To set the way a filter displays, make sure your dashboard is in edit mode and select the style of display in the Location field of the filter configuration window. Dashboard filters can be displayed in the following ways:

• Inline: The filter is displayed directly in the top bar of the dashboard.
  

  

• Popover: A summary value appears in the top bar of the dashboard; click the value to see the full filter.

  

• Placed behind a More button: A More button appears in the top bar of the dashboard with an indicator showing the number of filters it contains; click the button to see the filters and their values.

  

For most filter controls, you can choose to display them in inline or popover orientation through the Display field in the filter configuration window.

However, some filter controls can be displayed only in inline orientation; when you view a dashboard with these filter controls, you won’t see a Display field in the filter configuration window. These filter controls are:

• Drop-down Menu
• Single Day
• Timeframes
• Date Range

Additionally, some filter controls can be displayed only in popover orientation; when you view a dashboard with these filter controls, you won’t see a Display field in the filter configuration window. These filter controls are:

• Tag List
• Advanced

You can place any type of filter behind the More button by dragging the filter into the button after the filter is created. If all filters are displayed behind a More button, the button will instead show the word Filters.

Generally, we recommend that important or frequently used filters be displayed as inline or popover, and less frequently used filters be displayed as overflow.

Requiring a filter value

To limit dashboard viewers’ data access, use the access_filter LookML parameter instead of required filters.

By default, filters do not require values. If a filter that does not require a value is left blank, the data simply isn’t restricted by the filter field. For example, if you have a filter on a State field on a dashboard and that filter was not given a value, the dashboard returns data for all states.

If a filter does require a value, the dashboard won’t run until you select a filter value.

Additionally, you cannot create a dashboard schedule without selecting a value for a required filter or clear cache and refresh a tile that is linked to a required filter that does not have a value.

To require that a user enter a value in a filter before they can run the dashboard:

1. Make sure the dashboard is in edit mode.
2. Open the filter configuration window for the filter, either by adding a new dashboard filter, or by editing an existing dashboard filter.
3. Click the arrow to the left of the Additional options heading to reveal the additional options.
4. Select the Require a filter value checkbox.
   
   
5. Click Update to save your change.

The Require a filter value checkbox simply requires that viewers select any value. To restrict the values a viewer is able to select, set the allowed values in the Values field of the filter configuration window.

Required filters are helpful in limiting the sizes of queries sent to your database. To limit dashboard viewers’ data access, use the access_filter LookML parameter instead.

Setting up linked filters

In legacy dashboards, this feature is called faceted filters.

Dashboard filters can be linked so that the filter value options for one filter are narrowed based on the filter value or values selected for a different filter on the same dashboard.

For example, you can link a dashboard filter for City to a separate filter for State. The filter value options for City will change based on the state or states selected in the State filter.

Once you select a filter value for State, a link icon appears above the City filter. If you hover over the link icon, a box explains that the value options for that filter have been narrowed by the State filter.

You can go further and link a Zip filter to the City and State filters, narrowing the value options for Zip depending on the city and state selected:

If you link filters and your data contains no values for a “child” filter (the filter to be narrowed), given the values selected in the “parent” filter (the filter that narrows the child filter’s options), viewers will see a No matches found message for the child filter.

Linked filters work in one direction, from parent to child. If you make a selection in a child filter, it will not affect the filter value options presented to you for the parent filter.

How to link filters

When you are linking filters, any filter can be used as the “parent” filter (the filter that narrows another filter’s options), but the “child” filter (the filter to be narrowed) can only use a field of type string, commonly used for words or phrases, or type zipcode, commonly used for zip codes.

If there is only one filter on the dashboard, the ability to link filters is disabled.

To link one filter to another:

1. Make sure the dashboard is in edit mode.
2. Open the filter configuration window for the “parent” filter, either by adding a new dashboard filter, or by editing an existing dashboard filter.
3. Click the arrow to the left of the Additional options heading to reveal the additional options.
4. Select the Select filters to update when this filter changes checkbox.
5. Click the field box to reveal a drop-down with the other string and zipcode filters present on the dashboard.
6. Select the string or zipcode filter or filters you would like to link to this filter.
   
   
7. Click Update to save your change.

Moving and repositioning dashboard filters

Once filters are placed on a dashboard, you can move or reposition them by:

1. Entering edit mode on the dashboard.
2. Hovering over the filter to display the filter’s six-dot icon. The six-dot icon appears at the top left of the filter.
   
   
3. Clicking and holding on the six-dot icon, and dragging the filter to the position you want.

This drag and drop method can be used to reposition inline and popover filters, reconfigure which filters are placed behind a More button, and reposition filters within the More button drop-down.

If no filters are currently behind the More button, an empty More button temporarily appears while you are dragging a filter. You can move the filter over the More button to move it into the More button. If you do not place the filter in the More button, it disappears once you release your click.

If cross-filtering is enabled on a dashboard, you cannot reposition cross-filters using drag and drop, and entering edit mode clears any cross-filters currently on the dashboard.

Editing dashboard filters

In edit mode, a three-dot Filter Menu icon appears next to every filter.

To edit a filter, click the three-dot Filter Menu icon. This opens the filter menu, which displays options to edit or delete a filter:

Select the Edit menu option to open a filter configuration window, similar to the one used to add a filter, populated with the existing configuration for the filter. From there, you can update the filter-by field, the filter’s settings, or the tiles to update. Select Update or Cancel in the filter configuration window to save or cancel your changes.

Once you’ve edited filters, click the dashboard’s Update button to apply the new filter settings to the tiles.

Click Save in the blue toolbar to save your changes and exit edit mode.

Deleting dashboard filters

In edit mode, three-dot Filter Menu icon appears next to every filter.

To delete a filter:

1. Click the three-dot Filter Menu icon.
2. This opens the filter menu, which displays options to edit or delete a filter:
3. Select the Delete menu option to delete the filter.
4. Click Save in the blue toolbar to save your changes and exit edit mode.
5. Once you’ve deleted the filters, click the dashboard’s Update button to apply the new filter settings to the tiles.

Alternatively, you can select the Edit menu option, which opens the filter configuration window. Click the Delete button within the filter configuration window to delete your filter.

Deleting a filter cannot be undone.