Two tables are currently shared via Delta Sharing: platform_aggregated_usage, which holds aggregated client space and workspace data, and platform_detail_usage, which holds per-user activity. Each of their fields is described below.
To get access, please contact your account manager.
How the data is structured
Both tables contain flat data: each row corresponds to the finest level available, i.e. an object type (combined with a module, a workspace and a day, and for table 2, with a user). Metrics are calculated every day for the previous day (D-1).
Each field belongs to one of the three categories below, shown in the Type column of the tables:
| Type | Meaning and best practice |
|---|---|
| Dimension | Analysis axis (date, client space, workspace, module, object type, user…) used to filter or group metrics. |
| Detailed metric | Value specific to each row (object type). It can be summed and analyzed along all dimensions, with no specific handling required. |
| Global metric | Value calculated at a higher level (client space, instance or user) and repeated identically on every row of that level. Do not sum it across object types: use an average, a median, a maximum value, or filter on a single object type. The relevant analysis dimensions are given in the description of each metric. |
Table 1: General information (platform_aggregated_usage)
This table contains aggregated data at client space and workspace level. It covers the entire history and is updated daily with the previous day's metrics (D-1). It is suited to tracking platform adoption over time: licenses, active users, and object volume and documentation.
| Field name | Type | Description |
|---|---|---|
time_period | Dimension | Period considered for data collection: daily. |
instance | Dimension | Code of the instance where the client is deployed. |
clientspace_name | Dimension | Name of a company's client space. |
clientspace_uid | Dimension | Unique identifier of a client space. |
workspace_name | Dimension | Name of the workspace belonging to a client space. |
workspace_uid | Dimension | Unique identifier of a workspace. |
workspace_type | Dimension | Workspace classification: DEV (workspace used for testing purposes, not yet published) or PROD (workspace published and available to readers). |
module | Dimension | Platform module. |
obj_type | Dimension | Object type existing on the platform. |
nb_worskpaces | Global metric | Number of workspaces in the client space during the period considered. Possible analysis dimensions: time + client space / instance. |
assigned_stewards | Global metric | Number of Steward licenses assigned in the client space during the period considered. Possible analysis dimensions: time + client space / instance. |
assigned_readers | Global metric | Number of Reader licenses assigned in the client space during the period considered. Possible analysis dimensions: time + client space / instance. |
available_stewards | Global metric | Number of Steward licenses available in the client space during the period considered. Possible analysis dimensions: time + client space / instance. |
available_readers | Global metric | Number of Reader licenses available in the client space during the period considered. Possible analysis dimensions: time + client space / instance. |
active_stewards | Global metric | Number of active Stewards in the client space during the period considered. An active Steward is a Steward who logged in and created or modified at least one object. Possible analysis dimensions: time + client space / instance. |
active_readers | Global metric | Number of active Readers in the client space during the period considered. An active Reader is a Reader who logged in and performed at least one click. Possible analysis dimensions: time + client space / instance. |
plugins_activated | Global metric | Number of DataGalaxy browser plugins activated in the client space during the period considered. |
obj_created | Detailed metric | Number of objects created during the period considered, for the module, object type and workspace considered. |
obj_deleted | Detailed metric | Number of objects deleted during the period considered, for the module, object type and workspace considered. |
descr_added | Detailed metric | Number of descriptions added during the period considered, for the module, object type and workspace considered. |
links_added | Detailed metric | Number of links added during the period considered, for the module, object type and workspace considered. |
obj_viewed | Detailed metric | Number of objects viewed during the period considered, for the module, object type and workspace considered. |
obj_total | Detailed metric | Total number of existing objects for the object type considered. |
obj_descr | Detailed metric | Number of objects with a non-empty description. |
obj_descr_validated | Detailed metric | Number of objects with a non-empty description and a validated status. |
Table 2: User activities (platform_detail_usage)
This table contains non-aggregated data at individual user level. It allows a detailed analysis of each user's activity: logins, creations, modifications, views, etc. It is updated daily with the previous day's metrics (D-1).
User activity data is retained for a rolling 3-month period. If you want to keep a longer history, we recommend storing this data on your side.
| Field name | Type | Description |
|---|---|---|
time_period | Dimension | Period considered for data collection: daily. |
instance | Dimension | Code of the instance where the client is deployed. |
clientspace_name | Dimension | Name of a company's client space. |
workspace_name | Dimension | Name of the workspace belonging to a client space. |
workspace_uid | Dimension | Unique identifier of a workspace. |
workspace_version_uid | Dimension | Unique identifier of a workspace version. |
workspace_version_name | Dimension | Name of the workspace version. |
workspace_version_status | Dimension | Status of the workspace version. |
module | Dimension | Platform module. |
obj_type | Dimension | Object type existing on the platform. |
user_uid | Dimension | Unique identifier of a user. |
user_name | Dimension | User's first and last name. |
user_role | Dimension | User's role, as defined on the platform. |
user_team | Dimension | List of teams the user is assigned to. |
license | Dimension | User's license type. |
is_token | Dimension | Boolean indicating whether the user is an integration token or not. Useful in particular to exclude automated imports from the analysis of human activity. |
number_connections | Global metric | Number of times the user logged in to the platform during the period considered. Possible analysis dimensions: time + user + client space / instance. |
is_newcomer | Global metric | Indicates whether the user is new (had never logged in before). This metric equals 1 (= true) only on the day of the user's first login. Possible analysis dimensions: time + user + client space / instance. |
obj_created_user | Detailed metric | Number of objects created by the user. |
obj_modified_user | Detailed metric | Number of objects modified by the user. |
obj_deleted_user | Detailed metric | Number of objects deleted by the user. |
obj_imported_user | Detailed metric | Number of objects imported by the user. |
obj_owned_user | Detailed metric | Number of objects owned by the user. |
link_created_user | Detailed metric | Number of links created by the user. |
obj_viewed | Detailed metric | Number of distinct objects viewed by the user. |
nb_views | Detailed metric | Total number of views (a user can view the same object several times). |
How to handle large data volumes
The platform_detail_usage table can become very large for clients with many users. For example, with the 4 default modules (56 object types) over about 92 days, multiplied by the number of users, a client space with more than 240 users quickly exceeds 1 million rows.
- Prefer a workspace-level token. A client-space-level token multiplies the data by the number of workspaces.
- Above 240 users, the default limit of the Power BI Delta Sharing connector (1 million rows) may be reached. In that case, we recommend increasing the maximum number of rows extracted in the Power BI connector settings, or using another extraction method: