Hi community!
Usage data for all usage-based pricing meters is now available in Data Share. These new tables provide visibility into how your organization uses Rovo credits, Automation steps, Assets objects, Bitbucket, and AI agent resolutions in Customer Service Management. Data is at usage event level.
These are the 8 new usage tables in Data Share:
usage_tracking_platform_ai_credit – Rovo credit usageusage_tracking_platform_automation_credit – Automation usage, measured in stepsusage_tracking_platform_csm_resolution – Conversations resolved by Customer Service Management AI agentsusage_tracking_platform_assets_object – Assets object countusage_tracking_platform_bitbucket_build_time – Bitbucket Pipelines build time, in secondsusage_tracking_platform_bitbucket_lfs_storage – Bitbucket Git LFS storage, in bytesusage_tracking_platform_bitbucket_package_network – Bitbucket Packages network transfer, in bytesusage_tracking_platform_bitbucket_package_storage – Bitbucket Packages storage, in bytes
For table details, see the data share schema for organization data documentation.
The new tables will be gradually rolled out in the coming 2 days.
Where to find the new tables
If your organization already uses data share, the new schema ‘Usage' will automatically be available under Organization Schema. Organization admins can edit existing data share connections to add them to your existing data share.
If you are new to Data Share and on enterprise edition with Atlassian Analytics, here is a video walkthrough demonstrating how to set it up. Please turn to Data Share documentation for set up details
https://support.atlassian.com/analytics/docs/what-is-a-data-share/
If you do not have the Insights App or Atlassian Analytics yet and on enterprise edition, please contact support to get started.
How to use the new data
Columns shared by every usage table
usage_id – the unique identifier (primary key) of each usage record.org_id – the organization the usage belongs to.usage_count and unit_of_measure – the amount of usage in the record, and its unit (for example steps, resolutions, objects, seconds or bytes). Always check the unit before adding up or comparing values. The ai_credit table has no unit column, usage_count is the number of rovo credits.usage_event_datetime vs usage_timestamp – both are when the usage happened. usage_event_datetime is a UTC string to the minute, useful for grouping by time. usage_timestamp is a timestamp with full precision, best for filtering and sorting.producer_id – the ARI of the product site, Bitbucket workspace, entitlement or organization that produced the usage.
Gauge meters vs counter meters
The tables record usage in two different ways, which changes how you should aggregate them:
- Counter meters – each record represents incremental usage. To calculate total usage for a period, sum
usage_count: Rovo credits, Automation steps, Customer Service Management AI agent resolutions, Bitbucket build minutes and Bitbucket Packages network. - Gauge meters – each record represents the current total at that point in time, not an increase. Use the latest record, or an average or maximum, for a period: Asset objects, Bitbucket Git LFS storage and Bitbucket Packages storage.
Attribute usage to users, spaces and schemas
Who used it – actor_id links usage records to the atlassian_account table. In the Customer Service Management table, anonymous help seekers have an actor_id that starts with unidentified- and doesn't link to an Atlassian account.
Automation by space – container_id is the ARI of where the rule is scoped. When it's a Jira space (ari:cloud:jira:{cloudId}:project/{projectId}), join it to envelope_fields_resource_ariin the jira_project table in the data share schema for Jira family of products. Confluence spaces don't have a matching table yet. flow_id identifies the automation flow, which also doesn't have a matching table yet.
Assets by schema – schema_id joins to object_schema_id in the asset_object_schema table in the data share schema for Assets. A value of ALL means it covers all schemas.
Bitbucket by entitlement – all four Bitbucket tables include entitlement_id, the ARI of the commerce entitlement the usage belongs to.
⚠️Caveats
- Older records in all usage tables – Usage data is supported from June 1, 2026. Records prior to this date aren't supported and should be ignored.
feature column in the ai_credit table – This column is only valid for usage on or after September 1, 2026. Records before this date may show feature values that are no longer supported. When analyzing by feature, filter on CAST(usage_timestamp AS DATE) >= '2026-09-01'. - Automation table
producer_id – Older records may contain a product label, such as jira, instead of an ARI. - Site Name is currently not available in data share to join. You can use Atlassian API to retrieve site name.
forge_app_ref column in ai_credit is currently NULL
Send us your feedback
Comment below or contact support if you have any questions or concerns.
For schema questions, please refer to data share schema for organization data.
Thank you!