| Data stream | A connection that brings one source's data into Data Cloud, through a connector or the Ingestion API. | The Ingestion API takes streaming updates as they happen and bulk CSV loads for periodic syncs, so each source can use the speed it needs. |
|---|
| Data Lake Object (DLO) | Where ingested data is stored, in the source's original schema. | Keeping the raw shape means mapping mistakes can be fixed without re-ingesting. |
|---|
| Data Model Object (DMO) | An object in the Customer 360 Data Model that DLO fields are mapped to, such as Individual or Contact Point Email. | Only mapped fields can be used in segments and activations; unmapped data is invisible to the rest of the platform. |
|---|
| Identity resolution ruleset | Match rules that decide which records are the same person or account, and reconciliation rules that pick one value per field. | Scheduled runs happen at most once a day; real-time resolution uses exact matching through a real-time data graph. |
|---|
| Unified profile | The single customer record a ruleset produces, linked back to every source record. | It is what every team and every AI feature reads. Source records stay intact. |
|---|
| Calculated insight | A batch metric across profiles, such as lifetime value or spend rank. | Usable in segmentation and activation, but computed on a schedule rather than per event. |
|---|
| Streaming insight | An aggregate over web and mobile engagement data in rolling windows as small as one minute. | Drives data actions while the customer is still active. It cannot be used in segmentation or activation. |
|---|
| Data graph | A precomputed, read-only JSON record combining a customer's fields from several objects. | Real-time data graphs refresh instantly for active sessions and power real-time identity resolution, insights and segmentation. |
|---|
| Segment | An audience defined on unified profiles and insights. | Standard segments publish every 12 to 24 hours and rapid segments every 1 to 4 hours, so segments suit audiences rather than in-the-moment reactions. |
|---|
| Activation | Publishing a segment to a target such as Marketing Cloud Engagement, Marketing Cloud Personalization, B2C Commerce, cloud storage or an ad platform. | Activation is how a unified audience leaves Data Cloud to be used in campaigns. |
|---|
| Data action | An event Data Cloud sends when data changes, to a Platform Event, a webhook, Marketing Cloud Engagement or Amazon Kinesis. | The near-real-time path: a cart abandonment can start a journey or a repeated support visit can open a case. |
|---|
| Data space | A logical partition of data, insights and profiles, for example by brand or region, controlled with permission sets. | Keeps brands or regions apart without separate orgs. Requires the Data Spaces add-on. |
|---|