What is the difference between "metadata" and "in-app data"?
Atlassian distinguishes between "metadata" and "in-app data" based on the nature of the information and how it is utilized to refine platform performance and AI features. Both categories are de-identified and aggregated before being used to improve services for all customers.
Metadata consists of statistical characteristics, numeric fields, and derivatives of your data that describe how you use the platform without revealing specific identifying content. These attributes are used to optimize tools, improve search relevance, and provide root-cause analysis recommendations.
- Content attributes: Statistical descriptors such as readability scores of Confluence pages, complexity metrics, or semantic similarity scores between documents.
- System metrics: Numeric data entered into Atlassian-managed fields, including sprint end dates, story points, or Service Level Agreement (SLA) durations.
- Common patterns: Aggregated trends derived from search queries, Rovo Chat prompts, and common configuration values (e.g., custom work item types like "bug" or "repair ticket").
In-app data refers to the actual content generated by users while working within Atlassian applications. This category represents the specific information you create, store, and manage within the environment.
- Examples: Titles and body text of Confluence pages, descriptions and comments within Jira issues, custom status names, and workflow labels.
- Governance: Unlike most metadata, in-app data contribution settings can be toggled by organization admins.
- Privacy: Before being used for product improvement, this data undergoes strict de-identification to ensure no individual identities or sensitive personal information are exposed.
