Data Streams and DMO Mapping

This guide covers deploying web and mobile data streams in Data 360 and mapping them to Data Model Objects (DMOs). It applies to implementations using a web connector, a mobile SDK connector, or both.

Overview

  • Data Streams — Each connector (web or mobile) produces data streams based on its respective schema. These streams must be deployed and mapped to DMOs before personalization can function.
  • Data Lake Objects (DLOs) — Intermediate objects that hold raw ingested event data before mapping.
  • Data Model Objects (DMOs) — The structured objects that Personalization reads at decision time. DLOs must be mapped to DMOs to make data usable.

[ACTION REQUIRED: Review Note] - NOTE: The sitemap (web) or event tracking (mobile) should be built and validated before deploying data streams. Streams should be deployed progressively as schema objects are added — deployment should not wait until all events are finalised.


Best Practices

  • Data streams should be deployed progressively as schema objects are added — deployment should not wait for a complete schema.
  • Partial refresh mode must be used for all web and mobile data streams.
  • For web implementations with standard interaction names, the preconfigured event type derivation should be understood before DMO mapping is planned.
  • masterLabel values in the schema should be aligned with DMO field labels to maximise automated mapping coverage and reduce manual effort.
  • During validation, flat Data 360 events should be confirmed to include all mandatory fields with data types consistent with the schema.