AI SMS OpenAI Booking Assistant for Quote, Scheduling
Missed calls = missed revenue. In this post, we show how an AI SMS assistant—built with n8n and OpenAI—quotes jobs, checks availability, and books appointments directly into your calendar. With human-in-the-loop approvals and full audit logs, you get speed, control, and cleaner ops.

Project Title
Scrub Ducts — AI SMS Booking Assistant (n8n + OpenAI) for Quote, Scheduling & Audit Trail
Industry
Home Services / HVAC & IAQ (Air Duct & Dryer Vent Cleaning)
Technologies & Tools
n8n workflow orchestrator, OpenAI Chat (policy-first prompts + short-term memory), Twilio (SMS/MMS/voice), Wix Bookings/Contacts/Forms APIs, Gmail (human-in-the-loop approvals), Google Sheets (conversations/approvals/critique/opt-out logs)
Client
Scrub Ducts — Kentucky-based residential & commercial air-duct and dryer-vent cleaning provider focused on indoor air quality and operational transparency.
Challenges
Missed calls → missed revenue: Peak-season volumes caused delayed responses and lost bookings.
Inconsistent quoting: Prices and time windows varied by rep and time of day.
Tool sprawl: Availability lived in Wix; texting in Twilio; notes in email—no unified trail.
Compliance & brand tone: Needed SMS opt-out handling and on-brand language with oversight.
Solution
AI SMS Booking Assistant that understands intent, presents policy-compliant quotes and available time windows, collects required details, books into Wix, and confirms via SMS—with a human-in-the-loop (HITL) approval lane for sensitive replies.
Architecture (simplified):
Customer SMS/MMS/voicemail → Twilio → n8n inbound → OpenAI policy layer + short-term memory → Wix availability → contact find/create → form submit & confirm → booking create & confirm → compose reply → (optional HITL via Gmail) → send SMS → Google Sheets logging (convos/approvals/critique/opt-outs)
Key design choices:
Policy-first prompts: Strict quoting & booking rules; one-booking-per-thread idempotency.
HITL safety: Approve/revise/manual follow-up path for edge cases.
Observability by default: Every message and decision logged to Sheets for QA and training.
Consent & compliance: STOP/START updates an opt-out registry automatically.
Early Indicators & Next Steps
Response time: First reply now measured in seconds, not minutes/hours.
Operational consistency: Quotes/time windows sourced from APIs and policy—reduced variance.
Audit trail: Full conversation and approval history available for QA and coaching.
Next steps:
Slot ranking using drive time/tech skills.
Web deep-links for complex, multi-service jobs.
Weekly prompt/policy reviews using critique logs.
BI dashboards (Looker/Power BI) on top of Sheets data.
""
-
