This page lists terms used in the Looker product and user documentation.
$ is a substitution operator that lets you reference previously defined objects in LookML.
see also substitution operator
- access, access level
Admins have a variety of options for limiting what users can view and interact with in Looker.
see also content access, data access, feature access
A data action lets users perform tasks in other tools, directly from Looker. For example, the action can cause an email to be sent or can set values in other applications — or it can do anything else that you can configure a receiving server to do. The receiving server must be able to accept a JSON POST.
- Action Hub
Looker’s Action Hub is a multi-tenant service that forwards data to Looker’s integrated services. Any data your users send using an action will be processed temporarily on the Action Hub server rather than in your Looker instance. When your Looker admin or developer has set up an integrated service and a field tag to provide access to that service, you can select that service as an action when you drill into data and interact with that service.
see also action
- Application Time Zone
Application Time Zone is an admin setting for the default time zone in which scheduled Looks and queries run, where supported. When User Specific Time Zones are enabled, the Application Time Zone becomes the default time zone for users who do not have a time zone value set for their accounts.
see also time zone settings, User Specific Time Zones
- asynchronous query
An asynchronous (or async) query is a data request that makes one call to start the request, one or more calls to check the completion status of the query, and one call to fetch the results of the completed query. Async queries can help avoid freezing apps, connection timeouts and long dashboard load times.
- block, Looker Block
Looker Blocks are pre-built pieces of LookML that you can use and customize to your exact specification. From optimized SQL patterns to fully built-out data models, blocks can be used as a starting point for quick and flexible data modeling in Looker.
- browse, browsing
Browsing involves viewing, sharing, sending, and downloading data from dashboards, Looks, and Explores.
- closed system
Also called a multitenant installation, a closed system silos content to specific groups and prevents users from different groups from knowing about each other.
Formerly known as Discourse, the Looker Community is a user forum featuring posts, discussions, questions, and ideas shared among Looker users and experts.
In the Admin section of Looker, you establish the database connection from which a model will retrieve data.
Constants, defined with the LookML
constantparameter in a project manifest file, allow you to specify a value that can be reused throughout a project. You can reference constants anywhere in your model where strings are accepted.
In Looker’s documentation, the term content typically refers to Looks and dashboards.
- content access
Content access controls whether a user or group can view or make changes to a board, or to a folder (called a Space prior to Looker 6.20) and its contents. The two content access levels are View and Manage Access, Edit.
- Content Validator
Looker’s Content Validator searches your LookML for references to model, Explore, and field names. Developers can use the Content Validator after making changes to a project to check that their changes did not impact any of their users’ saved Looks or query-based dashboard tiles. It can also be used to find and replace LookML elements to fix errors due to changes.
- customer-hosted (deployment, instance, installation)
A “customer-hosted” deployment means that the product is installed by or for the customer at the customer’s premises or on a customer-controlled server within a third-party data center. A customer-hosted deployment includes the in-product services, meaning the services hosted by Looker and accessible through the product, specifically licensing data, configuration backups, system error reports, data actions, and support tickets, as further described in the Application Data Shared by Looker section of Looker’s security webpage. Looker support generally has no access to these instances for support or deployment purposes, and the customer must execute their own version updates.
see also Looker Cloud
A dashboard is essentially a collection of one or more saved queries, displayed as visualization or text tiles together on one page.
see also LookML dashboard, user-defined dashboard
- dashboard element, element
An element is a tile or visualization on a LookML dashboard, created using the
- dashboard file
A LookML dashboard is defined in a LookML project file with the extension
see also LookML dashboard
- data access
Data access controls which data a user or group is allowed to view. This type of access can be restricted or granted either at the user level or at the data level.
- Database Time Zone
Database Time Zone is an admin setting for the time zone Looker uses to interpret your raw data.
see also Query Time Zone, time zone settings
You can use one or more datagroup parameters to define a caching policy, to specify when to rebuild persistent derived tables (PDTs), and to trigger schedules.
- derived table
Derived tables allow you to create new tables that do not already exist in your database. A derived table is created in a Looker view file and allows the user to treat the output of any SQL query as if it were a table in the database (typically used for “fact” tables).
see also ephemeral derived table, native derived table, persistent derived table
- Development Mode
A developer can enter Development Mode to make and test LookML changes. The changes won’t affect other users until they are deployed to production.
The SQL “flavor” of a database. Examples of supported dialects are Amazon Redshift, PostgreSQL, MySQL, and Google BigQuery Standard SQL or Legacy SQL.
A dimension is a field that represents an attribute, fact, or value, which can be selected from the field picker within an Explore and can be used to filter a query. Common dimensions include such attributes as dates, names, and IDs, and often correspond to columns in your underlying data table. A dimension can also be created within a view file.
- dimension fill
Dimension fill is a feature that lets you instruct Looker to fill in missing dates or values for a given dimension, such as a date dimension with some years missing. You can avoid misleading graphs by preventing Looker from connecting the values in an incomplete set. The dimension fill option can be turned on or off with the
- dimension group
Using a dimension group, you can create multiple dimensions for a single underlying date or time column in the database. For example, you could split a
duration-type dimension group into intervals of days, weeks, months, and so on.
Looker makes it possible to drill into the data on a visualization to get more specific information about a specific data point. To do so, simply click on the part of the visualization about which you’d like more information.
Embedding involves using
iframe code to place an object (e.g., Looker charts or tables) into a website, spreadsheet, or other location outside of Looker. An embed user is a user who is interacting with a Looker object that is embedded in a location outside of Looker. Embedded content can be public or private (requiring either a Looker login or an SSO authentication).
A regular derived table — sometimes called an “ephemeral” derived table — is temporary and not written to your database.
see also derived table
An Explore is the starting point for queries. An Explore shows a specified set of fields from its associated view file, and these fields can be selected from the field picker to construct a query, which can be saved as a Look or dashboard tile. Explore URLs can also be shared.
Exploring involves using data to answer questions in Looker.
An Explore file is a LookML project file with the extension
.explore.lkml. Can be used for extending Explores across models and for defining native derived tables.
explore parameter adds a view to Looker’s menu of Explores. As a best practice, an Explore should be defined inside of a model file. Explores reference views and each Explore can contain joins to other views. An Explore can also be defined in an Explore file that is then included in a model file.
see also Explore file
- fanout (n.), fan out (v.)
A fanout occurs when one row of a primary data table can correspond to more than one row of a joined table, resulting in duplicated records and incorrectly calculated aggregations. In Looker, the fanout problem is avoided through the use of symmetric aggregates and by correctly defining the data set’s primary key.
- feature access
Feature access controls the types of actions a user is allowed to take in Looker. This type of access is managed by permission sets.
Explores and views contain fields, mostly dimensions and measures, which are the fundamental building blocks for Looker queries.
- field picker
The field picker displays the dimensions and measure applicable to the data shown in an Explore. The Looker developer or admin configures these dimension and measure options. The field picker may also display filter-only fields.
- filter expression
Filter expressions are an advanced way to filter Looker queries. You can use them in the Explore section of Looker by adding a filter and choosing the matches (advanced) option. They are also used in LookML for elements that take a
filterparameter. You can write filter expressions to filter on a string or to partially match strings, date and time, Boolean values, numbers, and location fields.
- In the Looker UI, a folder is a place where dashboards, Looks, and other folders (subfolders) are stored. Each user has a personal folder, and a Looker instance can also have various kinds of shared folders. Access to content in Looker is allocated at the folder level. Folders were called “Spaces” prior to Looker 6.20.
- In the Looker IDE, a folder is an organizational structure for your LookML files.
Looker functions let you transform your data or reference data in complex ways. They are similar in nature to Excel functions.
- group, user group
Users can be added to one or more groups. Groups are useful for managing users’ access to particular data, features, and content within Looker, as well as for assigning roles to users in bulk.
- Help Center
Looker’s Help Center contains articles written by Looker SMEs about specific use cases, best practices, and troubleshooting scenarios.
The server (or cluster of servers) that hosts Looker. Each client uses a production server (and, potentially, a staging server).
- (n.) The
joinparameter enables you to define the join relationship between an Explore and a view, so that you can combine data from multiple views. You can join in as many views as you like for any given Explore.
- (v.) Combine data from multiple views by defining the relationship between an Explore and a view through a
- (n.) The
- Labs feature
The Labs panel under the Admin menu lists Looker’s new, under-development features. These must be enabled by the Looker admin.
- legacy feature
Some deprecated product features have a legacy feature option that allows for the continued use of the deprecated feature for a period of time (which may be useful for developing and implementing a migration plan). The Admin section of Looker lists the features for which the legacy feature option can be turned on.
see also Looker expression
- Liquid parameter
In Looker, a Liquid parameter is a LookML parameter field that has been created using elements of the Liquid templating language.
- Liquid variable
A Liquid variable is an element of the Liquid templating language that can be used with LookML to create dynamic content.
- Look (n.)
A Look is a single table or visualization saved as its own individual report. Looks can be added to dashboards, scheduled, shared, and made public. Any changes made to a Look will be reflected in any dashboards that contain it.
- Looker API
The Looker API is a secure, “RESTful” application programming interface for managing your Looker instance and fetching data through the Looker data platform. You can write applications or automation scripts to provision new Looker user accounts, run queries, schedule reports, etc. Just about anything you can do in the Looker application you can do via the Looker API.
- Looker Cloud
Looker Cloud means the product is installed and provisioned to a customer on a web-connected platform that is run in a third-party hosting facility designated by Looker. For more information, see the Cloud Security Architecture section of Looker’s security webpage.
- Looker expression
Table calculations, custom fields, and custom filters rely on Looker expressions (Lexp) to perform calculations. A Looker expression is built from a combination of functions, operators, and fields, and possibly constants or variables.
LookML is a language for describing dimensions, aggregates, calculations and data relationships in a SQL database. The Looker app uses a model written in LookML to construct SQL queries against a particular database.
- LookML dashboard
A LookML dashboard is written entirely using LookML (as opposed to a user-defined dashboard, which is created by using the visualization editor).
see also user-defined dashboard
- manifest, manifest file
A manifest (or project manifest) is where you set project-level settings, like listing files from other projects to import into the current project, defining LookML constants, and specifying model localization settings.
A measure is a field in an Explore that represents measurable information about your data, such as sums, counts, and so forth. A measure is declared in a view file and can be of an aggregate or non-aggregate type.
- Merged Results
The Merged Results feature allows you to combine data from different Explores (even from different models, projects, or connections). Using the Merged Results feature, you can create a query from an Explore, then add queries from other Explores to display the merged results in a single table. The Merged Results feature performs similarly to a left join in SQL: it’s as if the added query is being left-joined into the primary query.
A model is the semantic layer in Looker that controls logic and gates data access for users. This is created by developers as a model file within the LookML project, and contains information about which tables to use and how they should be joined together. Multiple models can exist for the same database connection in a single LookML project, and each model can expose different data to different users.
- model file
Inside a LookML project, a model file specifies a single database connection, the set of Explores that use that connection, and the Explores themselves, as well as how they should be joined together.
- model set
A model set defines what data and LookML fields a user or group can see. An admin can select a combination of LookML models to which a user or group should have access. It must be used as part of a role to have any effect.
see also role
- native derived table
Native derived tables (NDTs) perform the same function as writing SQL-based derived tables, but are defined in LookML and refer to dimensions and measures in your model.
see also derived table
Admins can manage permissions to determine which users and groups can access content, data, and features. Permissions can be model-specific or instance-wide. Permission sets must be used as part of a role to have any effect.
- permission set
Permission sets are combined with model sets in a role.
see also role
- persistent derived table
A persistent derived table (PDT) is a derived table that is stored in the scratch schema of a database and can be regenerated on a schedule of your choosing. It can be referenced in a query without running the underlying SQL each time it is called.
see also derived table
- primary key
The primary key is the dimension that has exactly one unique value for each row of data. When data tables are joined together in a one-to-many relationship, the primary key must be defined correctly in order to avoid a fanout.
- primary query
A primary query is a single query created from a single Explore. When working with merged results, the primary query is a similar concept to the primary ID when joining multiple tables in SQL.
- Production Mode
In Production Mode, changes that have been made in Development Mode (but not pushed to production) are not yet reflected in the data model shared by all users. In Production Mode, LookML files are treated as read-only. A developer can enter Development Mode to make changes to LookML files and push those changes for others to see in Production.
- project, LookML project
In Looker, a project is a set of related models and other files (like Explores, views, and LookML dashboards) that you will use to define your data model. In general, a project corresponds to a single Git repository.
- Query Time Zone
Query Time Zone is an admin setting for the time zone in which to show dates and times when querying.
see also Database Time Zone, time zone settings
A role defines the privileges that a user or group will have for a specific set of models in Looker. Create a role by combining one model set with one permission set.
see also model set, permission set
- scratch schema
A scratch schema is a schema in the underlying database that is designated to store Looker PDTs.
see also persistent derived table
A Space is a folder where dashboards, Looks, and other Spaces (subspaces) are stored. Each user has a personal Space, and a Looker instance can also have a variety of shared Spaces. Access to content in Looker is allocated at the Space level. Spaces have been renamed folders as of Looker 6.20.
- SQL Runner
Accessible through the Develop menu or from an Explore (with the appropriate permissions), SQL Runner lets users run raw SQL against their allowed connections. SQL Runner can be used to perform database functions, create and add derived tables to projects, to leverage the
EXPLAINfunction, among other uses.
A subfolder is a folder that is stored within another folder. Folders were called Spaces prior to Looker 6.20.
A subparameter is a parameter that can be used under another (parent) parameter in order to further define, refine, or specify how that parameter functions.
A subspace is a Space that is stored within another Space. Spaces have been renamed folders as of Looker 6.20.
- substitution operator
The dollar sign (
$) is a substitution operator in Looker, and helps to make LookML code more reusable and modular. You can use this syntax to reference LookML objects that have already been defined.
- symmetric aggregates
Symmetric aggregates are functions in Looker that successfully return results for aggregations or measures. They help define the correct relationship between tables, thereby avoiding fanouts.
- system time zone
The system time zone is the time zone for which the server running Looker is configured.
see also time zone settings
- table calculation
Table calculations are similar to spreadsheet formulas and are performed on the results of a query, after the query has executed.
Themes are a way to customize the appearance of your embedded Looker dashboards. You can use themes to customize font family, text color, background color, and tile color.
Tiles are visualizations added to a dashboard from an Explore or a Look. Tiles can be query-based or Look-linked. Query tiles differ from Look-linked tiles because they are stored only on dashboards.
- time zone settings
Looker admins and users have a variety of options to convert and display time-based data in specific time zones.
see also Application Time Zone, Database Time Zone, Query Time Zone, system time zone, User Specific Time Zones
- User Specific Time Zones
User Specific Time Zones is an admin option that, when enabled, allows users to choose their own time zones. Queries will be created with the time zones of the users who created them.
see also Application Time Zone, Database Time Zone, Query Time Zone, time zone settings
- user-defined dashboard, UDD
User-defined dashboards are created by adding content through Looker’s user interface, rather than using LookML. This is the most common type of dashboard.
In Looker, a view can represent an underlying database table or a derived table. Views are the building blocks for Explores, which make the information in a view available for querying via the field picker in the Explore UI. By convention, a view is defined in a view file.
- view file
A view file is where you define the dimensions, measures, and other fields that are used in your LookML model.
see also model