Action Graph
How next best action is decided
How SignalFlow turns a recommendation into a governed, simulated execution plan. The flow is deterministic and consent first.
High intent purchase follow-up
Trigger: TRADE_REQUEST (intent 46, opportunity 99)
Send SMS
Open the conversation on the customer preferred channel.
Send Email
Fall back to email if there is no reply.
Create Task
This action is routed to a human for review.
Send SMS
Consent present, action allowed.
Send Email
Consent present, action allowed.
Create Task
This action is routed to a human for review.
Every workflow follows the same deterministic decision path.
New lead received
A signal enters the system and updates the customer intelligence graph.
Check consent
The policy layer evaluates channel consent, opt-out, and quiet hours.
Score intent
Signal context produces an intent score used to prioritize the action.
Choose channel
The graph selects SMS, email, voice, or a human task based on preference and policy.
Generate message
A draft is produced through the mock AI provider. Nothing is sent in Phase 0.
Queue follow-up
The action is queued as a simulated communication record.
Escalate if needed
High-value or sensitive cases route to a human for review.
Track outcome
Every step emits an audit event linking signal, decision, action, and outcome.