Skip to main content
id
string
required
name
string
required
description
string
required
Maximum string length: 10000
owner
string
required

The userId of the owner (or raw owner name/email for legacy records)

projects
string[]
required
tags
string[]
required
datasource
string
required
userIdTypes
string[]
required
sql
string
required
managedBy
enum<string>
required

Where this fact table must be managed from. If not set (empty string), it can be managed from anywhere.

Available options:
,
api,
admin
dateCreated
string<date-time>
required
dateUpdated
string<date-time>
required
ownerEmail
string

The email address of the owner, when the owner can be resolved to a known user.

userIdColumns
object

Maps an identifier type to the column holding it, for SQL that does not alias its columns to the identifier type names, e.g. {"user_id": "userId"}. May also be a single-level field path into a JSON column (properties.userId). Unmapped types use the identifier type name as the column name.

aggregatedFactTableSettings
object

Settings for maintaining shared daily aggregated tables (a subset of userIdTypes plus the daily update time and restate lookback window) used to speed up CUPED. Requires the data pipeline (pipeline-mode) feature.

timestampColumn
string

The column holding the event timestamp. Must be a date column on this fact table. Defaults to "timestamp" when unset.

eventName
string

The event name used in SQL template variables

columns
object[]

Array of column definitions for this fact table

columnsError
string | null

Error message if there was an issue parsing the SQL schema

columnRefreshPending
boolean

True while the fact table's column schema is being detected in the background. While true, columns may be empty or incomplete and metrics referencing not-yet-detected columns cannot be created.

archived
boolean
autoSliceUpdatesEnabled
boolean

Whether Auto Slice values for this fact table's columns are refreshed automatically in the background.