Salesforce Flow orchestration, subflows and fault paths
Reusable work and coordination
A subflow packages work that another flow calls. Define its input and output variables so the caller can use its result. A fault connector handles an operation that failed. Orchestration coordinates stages and assigned work; completing a stage is different from completing every possible business process outside it.
Salesforce’s stage-error guidance explains fault handling in orchestrations. Product availability and supported behavior depend on the Salesforce org.
Service-release example
A fictional service team needs a reviewer to approve a release before a background step updates its records. Keep the review work assigned until the intended reviewer completes it. If the update fails, inspect its error path and record which operation failed; a completed review doesn’t prove that the update succeeded.
A scheduled follow-up also needs a defined record and time. Test what happens when that record changes before the scheduled path runs. Don’t confuse this with a schedule-triggered flow that selects a set of records.
Modeled practice
Practice reusable flow contracts, fault recovery and record scheduled paths. The service-release project carries your configuration forward. These exercises use a modeled org with fictional data and don’t deploy to Salesforce.