The Operational Context
Fallon’s Home and Hearth manages recurring pellet deliveries and customer orders across multiple systems:
- Multi-Touch Pro (MTP / Odoo) for orders, customer information, inventory status, and lifecycle states
- Airtable for delivery scheduling (routes, loads, and drop sequencing)
- Ongoing phone and email interruptions during planning
Delivery planning required constant cross-checking across tools, sequencing loads, and adjusting routes in real time.
The Core Problem
Weekly delivery scheduling was a high-friction, interruption-heavy process. Planning required:
- Reviewing order records and customer details
- Confirming inventory status
- Updating routes, loads, and sequencing
- Responding to interruptions (phone/email) mid-planning
The result was ~10–15 hours per week spent on planning alone, with higher cognitive load, more context switching, and an increased risk of scheduling errors.
The System We Designed
Phase 1 — Delivery Logistics Assistant (FHH-1)
We built a workflow-specific assistant that integrates directly into existing systems and reduces manual coordination overhead.
- Integrated with MTP / Odoo (orders, inventory, lifecycle state)
- Integrated with Airtable (routes, loads, drop sequencing)
- Supports direct schedule updates with explicit human review + confirmation before changes
Phase 2 — Pellet Line Phone Assistant (FHH-2)
After internal planning was stabilized, we addressed inbound call friction with a production voice agent.
- Retrieves live inventory + pricing from MTP
- Performs delivery-status lookups against the scheduling database
- Escalates to humans when information is missing, inconsistent, or outside policy
Measured Impact
Internal Operations (FHH-1)
- Reduced planning time from ~10–15 hours/week to ~30 minutes or less
- Improved planning accuracy
- Fewer interruptions during scheduling
- More consistent execution of delivery workflows
Customer-Facing Operations (FHH-2)
- More accurate, consistent customer communication
- Reduced customer confusion around delivery status
- Deflection of routine inbound calls
- Live inventory/pricing visibility at time of inquiry
Note: Revenue/customer satisfaction metrics were not formally recorded in the engagement tracker. The case study focuses on documented operational outcomes.
Why It Worked
This engagement succeeded because the system was:
- Workflow-specific (built around real operational processes)
- Integrated (embedded into existing tools rather than replacing them)
- Controlled (human review gates before schedule changes)
- Stewarded (supported through ongoing engagement)
Applicability
This pattern applies to businesses that:
- Coordinate recurring deliveries or logistics
- Operate across multiple systems and databases
- Spend significant time on planning, scheduling, follow-up, or coordination
- Handle frequent inbound operational inquiries
Want to apply a similar system to your operations?
If you’re dealing with scheduling friction, follow-up gaps, or constant interruptions, we can help design a managed AI system that fits your existing tools.