Dynamic datastream configuration patterns

Two categories of patterns apply to Dynamic Datastream Configuration rule design: general rule-writing patterns that apply to all services, and Adobe Experience Platform-specific dataset strategy patterns. Each category addresses a different design decision and can be applied independently.

Before reading this page, review the event value taxonomy in the Dynamic Datastream Configuration overview. Classifying your events as Expendable, Analytical, or Actionable is a prerequisite for choosing the right pattern.

General patterns general-patterns

The following patterns apply to all services: Adobe Analytics, Adobe Target, Adobe Audience Manager, Event Forwarding, and Adobe Experience Platform.

Specific rules before broad rules granular-before-generic

Because the Edge Network uses first-match-wins evaluation, define more specific rules before broader ones. A generic rule placed before a specific rule matches first and the Edge Network never evaluates the specific rule.

Example: Place the more specific rule first, followed by the broader rule:

Write rules only for exceptions override-rules

Because unmatched events fall back to the default static datastream configuration, you do not need rules for your most common event path. Define rules only for the subset of events that should deviate from the default.

Example: A datastream with a primary non-profile-enabled dataset (Web Events - Analytics) and a secondary profile-enabled dataset (Web Events - Profile). Instead of writing rules for all event types, write a single rule to route Actionable events to Web Events - Profile. All other events automatically fall back to the primary dataset.

Experience Platform dataset patterns aep-patterns

When routing events to Adobe Experience Platform, choose a primary dataset strategy before writing rules. The strategy determines where unmatched events land, which is what happens to the majority of events in most implementations.

Both patterns support the same Edge Network services: Decision Management, Edge Segmentation, Personalization Destinations, and Adobe Journey Optimizer. See Experience Platform settings to enable them on your datastream.

Actionable first actionable-first

Actionable first means you prioritize profile ingestion, segmentation, and activation by ensuring every event goes to the Real-Time Customer Profile and all enabled Edge services unless a rule explicitly routes it elsewhere.

Set the primary (default) dataset to a profile-enabled dataset. Enable the Adobe Experience Platform Edge services you need.

All events go to the data lake, the Real-Time Customer Profile, and all enabled Edge services. Write rules to route Analytical events away from Profile and to disable Edge services for those events.

Analytical first analytical-first

Analytical first means you prioritize the data lake over the Real-Time Customer Profile by ensuring every event lands in a non-profile dataset unless a rule explicitly promotes it to the Real-Time Customer Profile and Edge services.

Set the primary (default) dataset to a non-profile-enabled dataset. Enable the Adobe Experience Platform Edge services you need.

All events go to the data lake only. Write rules to route Actionable events to a profile-enabled dataset and enable the appropriate Edge services for those events. Add rules to disable Edge services for Analytical events as needed.

When to choose Actionable first versus Analytical first which-pattern

Choose based on which event type makes up the majority of your traffic.

Rule of thumb: Write rules for the minority of your event types. If you have 3 Actionable event types and 15 Analytical event types, use Analytical first and write 3 rules to promote Actionable events to Profile. This keeps your rule count well within the 5-rule-per-service limit.

Next steps