GHL Conversation AI Agent

Bot Goals Architecture

Conversation AI in GHL is goal-driven, not script-driven. Each bot has:

Primary goal — what success looks like (booked call, qualified lead, form fill)
Guardrails — off-topic handling, objection responses
Handoff trigger — "book a call" intent → transfer to human / booking flow

Pipeline-Stage Filtering (the trick)

Don't run the bot on everyone. Filter by pipeline stage:

Trigger: Inbound SMS
  → Find Opportunity (by contact)
  → If/Else: Sales Pipeline stage = "Engaged"
    ├─ True: Conversation AI action
    └─ False: Do nothing (or route to sales)

Keeps the bot focused on the funnel stage where AI outperforms humans (tire-kicker nurture), skips hot leads who need a real rep.

Auto-Followup Config

Max touchpoints: 5
Max delay per touch: 24 hours
Escalating urgency tone across touches
Kill switch: any reply that includes "stop", "not interested", booking confirmation, or human-handoff keyword

Dead-Lead Revival

Run the bot as a one-shot blast on contacts with stage = Cold and last_contacted > 30d. Conservative messaging, no pressure. Recovers ~5-8% to booked calls on Iron Automations' own data.

Model Choice