Data Labels
Data labels are metadata that you can attach to data assets. Data labels help classify data and pass information to users. For example:
- One published data source might be considered more authoritative than other, similarly named ones. The certification data label can help you inform users which data source is recommended.
- A column in a database might contain outdated information. A warning data label can help you tell workbook authors and viewers that the data may not be up to date.
- A table of employee income might contain sensitive information that shouldn't be shared. A sensitivity data label can inform users that they should take care when using data from the table.
Note: Data labels are a more recent and extensible way of thinking about a way that Tableau classifies metadata. Certifications and data quality warnings, both of which were part of the Data Management license for Tableau Cloud and Tableau Server long before the term "data labels" existed, are now considered categories in the broader data label concept.
A Data Management license is required for all data label operations except for those related to the certification of published data sources.
Label categories
The data labels built into Tableau have one of the following categories: Certification, warning, or sensitivity.
Certifications
In a self-service environment with multiple publishers and a lot of assets, it can be difficult to find recommended content. Using certification, you can mark assets as trusted, and the assets display badges in various places across Tableau. For complete information, see Use Certification to Help Users Find Trusted Data.
Data quality warnings
Identifying problematic data is important for building trust with users. Data quality warnings allow you to mark data assets that have known issues. When you attach a data quality warning to an asset, a warning shows on it and any downstream assets that use it, making data consumers aware of problems with the source data. For example, if you mark a database table as deprecated, users viewing workbooks based on that table may see a warning. Furthermore, using the REST API, administrators can create new, customized data quality warning labels, adding nuance and specificity to the warnings that users can choose from. For complete information, see Set a Data Quality Warning
Sensitivity labels
Some data needs to be handled differently. Using Sensitivity labels, you can relay data sensitivity information to consumers of that data. When you mark an asset as sensitive, users browsing Tableau Cloud will see badges on it and any downstream assets that use it. For example, if you mark a database column as sensitive, a user browsing Tableau Cloud may see a "sensitive data" badge. Furthermore, using the REST API, administrators can create new, customized sensitivity labels, adding nuance and specificity to the range of classifications that users can choose from when using sensitivity labels.
Note: Sensitivity labels were introduced in Tableau Cloud June 2023. Earlier versions of Tableau Cloud and all versions of Tableau Server relay data sensitivity through the "Sensitive data" data quality warning instead of dedicated sensitivity labels.
For complete information, see Sensitivity Labels.
Assets you can label
Users can add labels to the following Tableau content and external assets:
- Databases
- Tables
- Columns (except for certification) (column labels introduced in Tableau Cloud October 2022 / Server 2022.3)
- Data sources
- Flows
- Virtual connections
- Virtual connection tables
Permissions required to interact with data labels on assets
Permissions required to view, add, update, and delete labels on assets are listed below:
- To view a data label, you must have read permissions on the associated asset.
- To add, update, or delete a data label other than a certification label, you must have write permission on the associated asset.
- To add, update, or delete a certification label, you must be an administrator, or else you must be a project leader or product owner for the project the asset is in.
- To add, update, or delete a certification label for an external asset not in a project, you must have the change permissions permission on the associated asset.
Customize labels
Beginning with Tableau Cloud June 2023, using the REST API, an administrator can customize built-in data labels, or create entirely new ones. This feature does not currently exist in Tableau Server. For information, see Customize Labels.
Comparison of data labels and tags
Tableau Cloud and Tableau Server also feature another solution for asset classification: Tags. Data labels and tags differ in significant ways:
| Area | Data Labels | Tags |
| Structure and control | Administrators control the range of data labels | No administrative control over the range of tags users add |
| Permissions | Ability to add/update/remove data labels controlled through asset permissions | Explorers and Creators can tag any assets that they can view |
| Appearance | Data label iconography is easily seen and color-coded by category and visibility level | Tags are generally less visible than data labels and have no iconography |
| Inheritance | Some data labels (like warnings and sensitivity labels) show on downstream assets | No inheritance |
| Searchability | Can be used as search filter criteria | Can be used as search filter criteria |
| API accessible | Access via REST API and Metadata API possible | Access via REST API and Metadata API possible |
| License requirements | Requires a Data Management license (except for the certification of published data sources) | No licensing requirements |
| Use | Structured categorization focused on providing important information that can influence users' use of data | Open ended method to categorize assets |
For more information on tags, see Use Tags(Link opens in a new window) in the Tableau Desktop and Web Authoring Help.