The First-Freeze Head Start System
Client identity withheld.
Results
$36K
Booked revenue in 90 days, at the shop's own prices
47
Tune-ups and service calls booked from past customers in October
38
Visits booked from forecast alerts before the freeze
44
Service calls booked by the phone assistant
29% → <2%
Calls left unanswered during freeze week
60% → 80%
October schedule capacity filled
Company Profile
Company
Residential HVAC shop. ~$880K revenue, 3 techs plus the owner, and a part-time CSR. ~1,500 past customers.
Stack
Entry-level field-service software, the business line forwarded to the owner's cell after hours, Google Local Services Ads.
Operating Need
Fill October and stay reachable when the first freeze hits, without adding staff.
The Challenge
October was always the shop's quietest month: techs on payroll, a board about 60% full. Then the first hard freeze came, call volume more than doubled in a few days, and the part-time CSR and the owner couldn't keep up. Many of those calls were furnaces that a fall check would have caught. The owner knew the pattern but never had time to get ahead of it.
The Audit
We reviewed two years of call volume by week, October capacity and the shop's service history.
Patterns Found
- October capacity averaged about 60% over two years
- About 29% of calls went unanswered during the first freeze week
- About 650 customers were overdue for a furnace check
- About 300 homes had furnaces 15+ years old
- After-hours calls during the freeze went to voicemail
- The shop had never contacted past customers before the season
The empty October and the overloaded freeze week were the same problem. The shop needed to pull the work forward.
The Architecture
Three systems working as six parts, sized for a four-truck shop.
Past-Customer Wake-Up
Cleaned 1,500 customers to 1,300 reachable numbers and texted overdue furnace checks first.
Capacity Matching
Sends only as many texts as a four-tech schedule can absorb in the next two weeks.
Weather Trigger
Watches the forecast and texts the oldest furnaces 5–7 days before a hard freeze.
Surge Warning
Alerts the owner to the expected call volume so he can plan on-call coverage.
Call Handling
An AI assistant in backup mode that picks up after 3 rings and after hours, follows the owner's no-heat emergency rules, and books routine work.
Monday Recovered Revenue Report
One page every Monday with bookings by source and running totals.
How the System Operates
The system books ahead and answers the rush. The owner sets the emergency rules and decides on-call coverage.
- Texts go out only in batches the schedule can absorb
- No-heat calls with vulnerable people at home always page the on-call tech
- Nothing goes out until the owner approves it in writing, and STOP replies are removed permanently
How It Flows
- 1Open slots and the forecast are checked daily
- 2The right customers are texted ahead of the cold
- 3Visits are booked, and freeze-week calls are answered and triaged
- 4Every booking is logged on the Monday report
A summary and recording of every after-hours call reaches the owner each morning.
Implementation
| Stage | Focus | Key Results |
|---|---|---|
| Stage 1 | Season audit | Freeze-week call pattern mapped; 650 overdue furnaces flagged |
| Stage 2 | Setup and approval | Customer list cleaned to 1,300 reachable numbers; messages and emergency rules approved |
| Stage 3 | October launch | Wake-up batches matched to openings; Call Handling live |
| Stage 4 | Forecast-driven | Weather trigger and surge warning live; Monday reports every week |
Ownership
The shop now has a repeatable fall plan sized to its own crew.
A customer list sorted by furnace age and last visit
Approved pre-season and forecast messages
Written emergency rules the AI follows
Every Monday report and call summary
Operational Impact
October now has paying work instead of open afternoons, and many furnaces get fixed before they fail. When the freeze arrives, the owner already knows it's coming, and every call is answered instead of going to voicemail. A four-truck shop now handles the season like a bigger one.