segmenting-datacloud

About segmenting-datacloud

A skill for the Segment phase of Salesforce Data Cloud, owning audience and insight work: creating and publishing segments, managing calculated insights, inspecting segment counts or membership, and troubleshooting audience SQL. It is scoped through the sf data360 segment and sf data360 calculated-insight commands and applies when the task is segment creation, publish workflows, member counts, or segment troubleshooting. It explicitly delegates adjacent work elsewhere: DMO, mapping, and identity-resolution tasks go to harmonizing-datacloud, downstream activation to activating-datacloud, read-only SQL and search-index queries to retrieving-datacloud, and Standard Data Model or session tracing to observing-agentforce.

Before acting, the skill gathers required context such as the target org alias, the unified DMO or base entity name, whether the user wants to create, publish, inspect, or troubleshoot, whether the asset is a segment or a calculated insight, and the expected success metric (member count, aggregate value, or publish status). Its operating rules treat Data Cloud segment SQL as distinct from CRM SOQL, run a shared readiness classifier from the orchestrating-datacloud skill (diagnose-org.mjs with --phase segment) before mutating assets, prefer reusable JSON definitions for repeatable creation, use --api-version 64.0 when segment creation is unstable on newer defaults, verify with counts or SQL after publish or run steps instead of assuming success, and prefer SQL joins over segment members when readable member details are needed.

The recommended workflow proceeds through classifying readiness, inspecting current segments and calculated insights, creating from JSON definitions, publishing or running explicitly, and verifying with counts or SQL. High-signal gotchas note that segment creation can require the 64.0 API version, that segment members returns opaque IDs, that publish or run can trigger asynchronous work even when a command returns quickly, and that an empty list usually means the module is reachable but unconfigured rather than unavailable. A defined output format reports the segment task, action, target org, artifacts, verification, and next step. The skill requires an external community sf data360 CLI plugin and a Data Cloud-enabled org, and points to shared definition templates and references in the orchestrating-datacloud skill.

FAQ

What tasks does this skill own versus delegate?

It owns segments, calculated insights, publish workflows, member counts, and segment SQL troubleshooting. It delegates DMO/mapping/identity-resolution to harmonizing-datacloud, activation to activating-datacloud, read-only SQL/search-index queries to retrieving-datacloud, and STDM/session tracing to observing-agentforce.

What does it require to run?

An external community sf data360 CLI plugin and a Data Cloud-enabled org. It also runs a shared readiness classifier (diagnose-org.mjs with --phase segment) from the orchestrating-datacloud skill before mutating audience assets.

Why might segment creation need a specific API version?

Segment creation behavior can be unstable on newer defaults, so the skill recommends passing --api-version 64.0 when creating segments to keep behavior reliable.

How do I get readable member details from a segment?

Use SQL joins rather than the segment members command, because segment members returns opaque IDs. Remember that Data Cloud segment SQL is distinct from CRM SOQL.

How should success be confirmed after publishing or running?

Verify with member counts or SQL queries instead of assuming success, since publish and run steps may kick off asynchronous work even when the command returns quickly.

All Files