Workflow Execution and Data Flow
The order of Workflow operation can be split into two concepts:
Workflow Execution
A Workflow processes Nodes in sequential order:
- Beginning at the root of the Node tree with either a "Start" or "Intercept" Node.
- Every subsequent node in the chain is processed in order. The execution may take the path of another branch based on conditionals.
- The Workflow ends either when it reaches an explicit "End" Node or when no more Nodes are available in the chain.

Workflow Data Flow
Each Node in a Workflow has an input and output data type.
With this typed data system, even though Nodes are processed sequentially, you do not need a direct line between two Nodes in order to pass data from one to another. Instead, data can be referenced using dot notation of a Node's alias and it's output alias:
$[node_alias].[output_alias]
TIP
You can view the data type by clicking on a Node and viewing the value within the parenthesis next to the name of the object.
