Skip to main content

Execution Graph

When a record is created, updated or deleted, Salesforce runs a chain of automations: flows, triggers, validation rules, Apex classes. The execution graph makes that chain visible for one object: its entry points, the business rules they reach, and the writes they trigger.

It is included on every plan.

Opening the graph​

  • From an object page in the portal, click View execution graph.
  • The graph can also be opened directly; you then choose the object in a searchable list.

The graph always refers to the version selected in the portal.

What the graph shows​

The graph starts from the object and groups its entry points by how they are triggered:

  • On create, On update, On delete
  • Via API
  • Scheduled jobs
  • Events
  • Asynchronous

Expand a group to see its entry points, then expand an entry point to follow what happens downstream. Nodes are of five kinds: Entry point, Business rule, Technical, Object and External system. Writes to other records appear as Write links.

Click a node to open its Node detail: type, event, execution phase, Apex class and method when relevant, and for a business rule its Condition (When) and Action (Then), with a link to its documentation page.

Controls​

  • Graph / Table -- switch between the visual graph and a table of the same nodes.
  • Filter -- show only some kinds of nodes.
  • Zoom in, Zoom out, Fit -- navigate large graphs.
  • Expand / Collapse / Collapse all -- control how much of the chain is displayed.

How it is built​

The graph is computed from your metadata, not guessed by AI. The order of execution, the phases and the links between components come from the Salesforce metadata and code of the scanned version. Running the same scan on the same org produces the same graph. Only the business wording of the rules is written by AI.

tip

The graph is the right tool for "which automation runs first?" questions -- for example when a flow and a trigger both update the same field.

Other diagrams in the documentation​

Beyond the object graph, the documentation includes diagrams for other kinds of components, for example:

  • Screen flows -- an interactive view of screens, decisions, actions and subflows, with Graph, Table and Journey views.
  • Functional processes -- an end-to-end view of the process phases. See Functional Processes.
  • Integrations -- the integration map of external systems.

Limitations​

  • The graph shows one object at a time. Business flows that span several objects are documented as functional processes.
  • It reflects the org at scan time. Changes made since are visible after the next scan.
  • If an object has no trigger, flow or automation, no graph is shown for it.
  • Paths decided at runtime by code (for example a class name computed dynamically) cannot always be followed; they are not invented.