All docs pages (49)
Process Graph
The automations Process Graph can read on an object (flows, triggers, workflow rules, validation rules), in the order they run, and fields more than one writes.
Updated 2026-06-08
For the setup screen (objects, Fast or Full, Customise), the Dashboard and Map, and six use cases with screenshots, read Process Graph: see every automation that runs when a record is saved.
When three flows, a trigger and two workflow rules all fire on the same object, the order they run in decides the outcome, and few people can recite that order from memory. Process Graph draws the automations it can read on an object, how they are wired, and where they may conflict.
The dependency graph
Process Graph draws the automations it reads on an object as an interactive graph:
- record-triggered and scheduled Flows
- Apex triggers
- Workflow rules
- Validation rules
- approval processes
Edges show what triggers what, so you can follow a save through the chain instead of imagining it.
What the scan reads. Only active flows are read. Fast mode reads up to 100 flows and 100 Apex classes; Full mode reads more. When a cap is reached, a banner says so and offers to switch to Full. Anything the scan did not read is not on the graph.
Execution order
The graph models Salesforce's order of execution, so you can see the sequence automations actually run in — before-save vs after-save, triggers vs flows vs workflow. This is where order-of-operations bugs hide: a before-save flow that sets a field, overwritten by an after-save trigger, producing a result that looks impossible until you see the order on screen.
Conflict and dead-code detection
Process Graph flags problems that cause hard automation bugs:
- Dead / unused (from version 1.1.13: No caller found) — automations that nothing the scan read starts or calls. Something outside the scan may still start them, so check before you delete one.
- Duplicate writers — multiple automations writing the same field. The classic source of "this field keeps changing and I do not know why."
- Object health scoring — a per-object signal for how tangled its automation has become.
Impact of an automation
Select an automation to see how many other automations and fields it affects downstream. Use it before you deactivate or delete something, together with your own testing.
Documentation export
Export the graph as Mermaid to put an automation diagram into a runbook or a pull request. It is generated from the org, so it matches what the scan read.
Related reading
- Flow Hub — best-practice checks for individual flows.
- Where Used and Object Data Dictionary (in the extension's Metadata menu) — where a field is referenced, and the object's fields.
