Labs overview

Looker Labs features are new, in-progress features, whose details may change over time. They are split into two categories:

• Beta: These features are expected to remain in the product, and errors are expected to be resolved at some point. However, these features may change in detail, and errors may not be fixed with the same speed as with normal features. You can see the list of current beta Labs features in Beta features, below.
• Experimental: These features may or may not remain in the product, and errors may or may not be corrected. These experimental features are intended to show you functionality that could be incorporated into Looker in the future and to get your feedback. You can see the list of current experimental Labs features in Experimental features, below.

You can choose to use these features or leave them disabled. You’ll see a list of these features on the Labs page in the General section of the admin menu. The list of available features may vary, so the feature names and descriptions are blurred in this example:

Each feature has a short description underneath it explaining the functionality that it adds to or changes in Looker. If you would like more details about a feature, check out the Looker Community or the Help Center, which may have more detailed descriptions, or reach out to Looker Support.

Enabling and disabling Labs features

To enable or disable a feature, click the switch next to the feature name.

Use caution when disabling a Labs feature. If users or developers have relied on the feature to create a certain behavior, disabling the feature will break that functionality.

Current Labs features

Looker supports the following beta and experimental Labs features.

Beta features

Beta Labs features are expected to remain in the product, and errors are expected to be resolved at some point. However, their details may change, and errors may not be fixed with the same speed as they would be for normal features.

Database Connection Pooling

Enables the use of preconfigured database connection pools on the Amazon Redshift, PostgreSQL, and Snowflake database dialects.

This option lets Looker use pools of connections through the JDBC driver. Database connection pooling enables faster query performance; a new query does not need to create a new database connection but can instead use an existing connection from the connection pool. The connection pooling capability ensures that a connection is cleaned up after a query execution and is available for reuse after the query execution ends. This Labs feature is disabled by default.

Extension Framework

The Extension Framework switch allows developers to build and run Looker-hosted applications as well as install and run extensions from the Looker Marketplace, such as the Data Dictionary. Please note that developing on the Looker extension framework is free to use while in beta and will remain free for developers indefinitely. Terms and support will be finalized and introduced when this feature reaches General Availability. This Labs feature is disabled by default.

Pricing, terms, and other details for Marketplace content and extensions such as the Data Dictionary are listed within the relevant store page. Looker admins must enable the Marketplace and Local Project Import Labs features to install and use Marketplace extensions.

IDE Metadata Sidebar

When it’s enabled, a sidebar in the IDE shows contextual LookML metadata about the model’s Explores and views, including where views are in used in Explores and joins, how Explores and views are extended, and how Explores and views are refined. This Labs feature is disabled by default.

Incremental PDTs

Enables Looker to append fresh data to a persistent derived table (PDT) or an aggregate table where an increment_key is defined instead of rebuilding the entire table. This Labs feature is disabled by default.

Marketplace

The Looker Marketplace is a central location for finding, deploying, and managing any type of Looker content, such as Looker Blocks, applications, visualizations, and plug-ins. You must enable the Local Project Import Labs feature to install packages from the Marketplace. This Labs feature is enabled by default.

New Query Streaming

Enables new, experimental query streaming, which may significantly improve query download performance. This Labs feature is disabled by default.

SQL Runner Vis

Adds the ability to visualize ad hoc queries from SQL Runner. This Labs feature is disabled by default.

Upgraded Explore Field Picker

Starting in Looker 7.20, the Upgraded Explore Field Picker Labs feature is enabled by default.

When this feature is enabled, an Explore will display an upgraded field picker with a smarter search capability, easy view and management of selected fields, Quick Start analyses for Explores with modeled queries, and other user experience improvements.

Visual Drilling

The Visual Drilling Labs feature is not supported by the new dashboard experience. Dashboards using the new dashboard experience support visual drilling through the use of the link parameter without the need to enable the Lab feature.

When Visual Drilling is turned off, the drill overlay always displays the data table.

This Labs feature is enabled by default. When Visual Drilling is enabled, the drill overlays for Looks, Explores, and legacy dashboards select the best visualization based on the underlying data, which may be a table visualization or some other visualization type. Or, with Visual Drilling enabled, a drill visualization can be customized by using the link parameter, as shown on the link parameter documentation page and in the More powerful data drilling Help Center article. When the visualization from a drill is not a table, buttons allow the user to switch between the default visualization type and a data table.

For more information about visualization types, see the Visualization types documentation page.

Experimental features

Experimental Labs features may or may not remain in the product, and errors may or may not be corrected. These experimental features are intended to show you functionality that could be incorporated into Looker in the future. We always welcome your feedback.

Custom Fields

This Labs feature is disabled by default. When it’s enabled, this feature lets users with the create_table_calculations permission create additional, temporary fields on the Explore page that (unlike table calculations) become part of the query running in the database. Other users can see the new fields if they have the create_table_calculations permission themselves, if they are given the Explore query’s URL, or if they are given access to a Look or dashboard that uses that field.

Use the Custom Fields Labs feature if you want to let some of your business users have these capabilities:

• Create semi-permanent measures and dimensions for a one-time or an infrequent analysis
• Create semi-permanent measures and dimensions without needing an analyst
• Visualize unmodeled data using Instant Explore from SQL Runner

For more information, see the Adding custom fields documentation page.

Dashboards in Drill Menus

Automatically suggests dashboards with matching filters.

Suggested dashboards will appear in the drill menu. In systems with many dashboards, this feature decreases dashboard performance.

Instant Dashboards

Dashboards display previous data while new queries are run.

If Instant Dashboards are enabled, when a dashboard loads, Looker pulls data from the cache to immediately display data from the previous run. The dashboard issues new queries that run in the background, and dashboard tiles are then updated as new data is retrieved. The spinner icon appears on the top right of each tile that is still loading, indicating that a query is running in the background and that the currently displayed data is from the last time the query was run. This has the advantage of users not seeing a blank dashboard while it’s loading, but the disadvantage of possibly showing stale data while the new queries run. Instant Dashboards are only supported by legacy dashboards.

Enabling Instant Dashboards also affects how long data is stored in the cache, since this feature relies on the existence of cached data to display.

Local Project Import

A LookML project can reference files in other locally maintained projects through the use of the local_dependency keyword in a manifest file. Enabling this Labs feature is not required to use the remote_dependency keyword. This Labs feature is disabled by default.

New Groups Page

The Groups page in the Users section of the Admin menu lists and lets you manage all the user groups on your Looker instance. This Labs feature is disabled by default. When New Groups Page is enabled, the Groups page displays a new, redesigned table and pagination that simplifies group management and increases performance on instances that have a large number of groups. For more information, see the New Groups page section of the Groups documentation page.

Reduce Filter Queries

Reduce the number of queries sent by filters. This moves any filter that requests suggestions into a popover where it will not fetch suggestions until a user clicks on it.

Render Long Tables

Reduces the dashboard PDF rendering times for dashboards containing tables with many rows, by displaying an experimental faster table.

This Labs feature defaults to off.

If Render Long Tables is enabled and the Expand Tables to Show All Rows download or schedule option is selected, dashboard tiles containing table visualizations may look slightly different in downloaded or delivered PDFs than they do inside Looker. The following differences may be noticeable in the PDF:

• Customizations to background colors and font sizes are removed from column headers and subtotal rows.
• Tables appear in the white theme.
• The sort icon does not appear on pivoted tables that were not manually sorted.
• Tables with Size Columns to Fit enabled stretch to the full width of the tile.

Trend Lines in Visualizations

Enables the addition of trend lines to Cartesian charts.

This feature enhances the visualization options available for column, bar, line, area, and scatter charts using a variety of regressions.