• Fonts and colors have been updated across the entire Looker application to align with Google branding. This can be enabled or disabled with the Legacy Branding toggle under Legacy Features. The toggle is disabled by default in Looker 7.20.
• For BigQuery Standard SQL connections, the Treat the Datatype Datetime as a Timestamp legacy feature is disabled by default.
• The Upgraded Explore Field Picker beta Labs feature is now enabled by default for all instances.
• When Looker initiates a Snowflake database connection, the AUTOCOMMIT parameter is now set to TRUE by default.
• Filter expressions can no longer be added to measures of type: number.
• Looker no longer double-encodes links that contain special characters.
• The Sandboxed Custom Visualizations Labs feature is now out of Labs and generally available.
• During startup, Looker now validates the dialect and version of the internal database. If something other than the default in-memory HyperSQL or a MySQLv5.7.x database is used as the internal database, Looker logs an error message before performing any startup migrations.

• The deployment manager allows developers to deploy any commit to production from any branch.
• The deployment manager separates deploying from merging: developers who don’t have the ability to deploy are now able to merge changes with this feature.
• This feature also adds Git tags from the UI to mark version numbers or other metadata and allows rolling back to prior versions.

• Looker’s new dashboard experience is out of Labs and generally available.
• The Cross-filtering Labs feature is enabled by default on dashboards. Note: This Labs feature is not available on legacy dashboards.
• Fonts and colors have been updated across the entire Looker application to align with Google branding. This can be enabled or disabled with the Legacy Branding toggle under Legacy Features. The toggle is enabled by default in Looker 7.18 and will be disabled by default in Looker 7.20.
• The Aster JDBC driver has been removed from Looker. All Aster connections will require a custom JDBC driver install.
• The SAP HANA JDBC driver has been upgraded to v2.5.46 to give users the ability to set the default schema, configure SSL, verify SSL connection options, and get additional support for the SAP HANA JDBC parameters.
• The Postgres Pre-9.1 SQL dialect is being downgraded to a support level of Integration. Postgres 9.5+ remains a supported dialect option.
• Using the Liquid variable _filters[] in a derived table’s SQL without the sql_quote parameter is now discouraged and will return a deprecation warning when the filter value is rendered in SQL. This warning is expected to be upgraded to an error by 2021.
• The LookML IDE will now render a warning if two joins without view_labels join the same labeled view, which would result in duplicated LookML fields.
• If the suggest_dimension parameter references an invalid or excluded field, the LookML IDE will render a new warning.
• If localization is used on a project and a dimension group with type: time is defined in the LookML, a LookML warning will be rendered. The warning will suggest splitting the dimension group into individual dimensions.

• Fonts and colors in the Looker platform have been updated to reflect GCP branding.
• Users can revert to the Looker branding with the Legacy Branding toggle, which is enabled by default in Looker 7.18 and will be disabled by default for all instances in Looker 7.20.

• Users will now be able to see data anytime and anywhere through the Looker mobile app.
• The mobile app is available to customers using Looker 7.16+.
• The app is available through the Google Play Store for Android 8+ devices and through the Apple App Store for iOS 13+ devices.

• Users will no longer have the ability to cross-filter on custom dimensions.
• Filtering by a field that is not a part of an Explore used on dashboards (beta) will display a banner with the message This filter was created using a filter by field outside of the Explores on this dashboard. You can modify the settings and tiles to update, but will need to create a new filter to change the field.
• Referencing a field with the syntax ${field_name} in a Custom Filter will cause the Liquid parameters {{ field_name._is_filtered }} and {{ field_name._in_query }} to evaluate as true.
• Values for hidden user attributes do not have to be re-entered when you update a Marketplace extension configuration or Marketplace extension version.
• User attributes referenced by access filters that have the values %, NULL (for String Filter [advanced]) and <0, >=0, NULL (for Number Filter [advanced]) will no longer automatically generate a WHERE clause.
• Users without the schedule_external_look_emails permission will no longer be able to save an alert if they add an email address that is not on the domain whitelist.
• The Load Assets from CDN setting has been removed from the Settings page in the General section of the Admin panel for Looker-hosted instances. This setting has not been removed from customer-hosted instances, unless the --force-cdn startup flag has been enabled on the instance.
• Save and Schedule menu items will no longer appear on embedded Explores when Remove Look Navigation is enabled.
• The create_sso_embed_url API endpoint is no longer metered as an admin call and will be excluded from the admin activity quota.
• The connections drop-down list on the Persistent Derived Tables page in the Admin panel will only list connections that have PDTs enabled.
• Adding the LookML parameter publish_as_db_view: yes to a PDT definition in production will now create a view without needing to trigger the table to rebuild or without having to modify the SQL or the LookML of the derived table.

• Developers now have the ability to use the following persist parameters inside the materialization block of an aggregate table: indexes, cluster_keys, partition_keys, distribution, distribution_style.
• Aggregate tables no longer need to be an exact timeframe match of the filters in an Explore query. For example, an aggregate table that has a daily timeframe dimension can be used for an Explore query that filters on week or month. As long as the dimension in the aggregate table has a timeframe with a finer (or equal) granularity as the timeframes of the dimensions and filters in the Explore query, the aggregate table can be used for the Explore query.

• Ensures that users can easily discover relevant and helpful marketplace choices.
• Enables Looker users to discover, install, and uninstall content like blocks, applications, and custom visualizations. This initial release of the marketplace is publishing the following:
  • Sales Application
  • 23 Looker Blocks
  • 12 custom visualizations

• For every rebuild of a PDT, the given table name will now be unique. This will yield increased stability and faster rebuilds for certain dialects.
• Stable, statically named view tables can now be generated on top of PDTs. A stable database view will only be published for its underlying PDT when that PDT gets rebuilt. By having a static, friendly name, the view table allows you to consistently access a PDT via SQL on the source database. This feature provides users with the ability to hook up other analytics or data science tools to Looker-governed data. Learn more about stable view tables.

Learn more about how boards can help your organization.