By Sagar Shankaran, Founder of CallSphere
HVAC companies use AI voice agents to run seasonal maintenance campaigns that fill schedules 6 weeks before peak season, eliminating the feast-or-famine cycle.
Key takeaways
The HVAC industry operates on one of the most punishing seasonal cycles in all of home services. Summer and winter bring a flood of emergency calls — broken air conditioners in July, failed furnaces in January — that overwhelm capacity. Spring and fall are dead zones where technicians sit idle and revenue craters.
The numbers illustrate the problem. A typical HVAC company with 12 technicians generates 70% of its annual revenue in just 5 months (June-August and December-January). During peak months, the company turns away 30-40% of service requests because every technician is booked. During off-peak months, technician utilization drops below 40%, and the company burns cash on payroll, truck leases, and insurance with insufficient revenue to cover costs.
The proven solution is proactive seasonal maintenance — spring AC tune-ups and fall furnace inspections. Maintenance agreements generate predictable recurring revenue, fill the shoulder-season schedule, and create a pipeline of equipment replacement opportunities. The problem is reaching customers at scale. An HVAC company with 5,000 past customers in its database might convert 15-20% to maintenance agreements if every customer were contacted personally. But calling 5,000 customers manually takes a team of 3-4 people working full-time for 6-8 weeks — time and labor that most HVAC companies simply do not have.
HVAC companies have tried every channel to drive seasonal maintenance bookings:
flowchart LR
CALLER(["Homeowner"])
subgraph TEL["Telephony"]
SIP["Twilio SIP and PSTN"]
end
subgraph BRAIN["Field Service AI Agent"]
STT["Streaming STT<br/>Deepgram or Whisper"]
NLU{"Intent and<br/>Entity Extraction"}
TOOLS["Tool Calls"]
TTS["Streaming TTS<br/>ElevenLabs or Rime"]
end
subgraph DATA["Live Data Plane"]
CRM[("CRM and Notes")]
CAL[("Calendar and<br/>Schedule")]
KB[("Knowledge Base<br/>and Policies")]
end
subgraph OUT["Outcomes"]
O1(["Service appointment booked"])
O2(["Quote sent via SMS"])
O3(["Tech dispatched today"])
end
CALLER --> SIP --> STT --> NLU
NLU -->|Lookup| TOOLS
TOOLS <--> CRM
TOOLS <--> CAL
TOOLS <--> KB
NLU --> TTS --> SIP --> CALLER
NLU -->|Resolved| O1
NLU -->|Schedule| O2
NLU -->|Escalate| O3
style CALLER fill:#f1f5f9,stroke:#64748b,color:#0f172a
style NLU fill:#4f46e5,stroke:#4338ca,color:#fff
style O1 fill:#059669,stroke:#047857,color:#fff
style O2 fill:#0ea5e9,stroke:#0369a1,color:#fff
style O3 fill:#f59e0b,stroke:#d97706,color:#1f2937
Direct mail postcards cost $0.75-1.25 per piece and generate a 1-3% response rate. For 5,000 customers, that is $3,750-$6,250 in postcard costs for 50-150 bookings. The cost per booking is $25-$125 — workable, but the volume is too low to fill a schedule.
Email campaigns are cheaper but perform worse. HVAC industry email open rates average 18-22%, with click-through rates of 1.5-2.5%. Many customer email addresses are outdated. The resulting 40-60 bookings from a 5,000-customer list barely makes a dent in the schedule.
Hear it before you finish reading
Talk to a live CallSphere AI voice agent for home services in your browser — 60 seconds, no signup.
Text message blasts risk TCPA violations if consent is not properly documented. Even with proper consent, text campaigns yield 3-5% booking rates — better than email, but still insufficient to fill 6 weeks of schedule capacity.
The phone call remains the highest-converting channel for maintenance agreement sales. A personal call to a past customer converts at 12-18% — 5-10x higher than any digital channel. The constraint has always been the cost and time required to make thousands of calls.
CallSphere's HVAC outbound campaign agent calls past customers with personalized maintenance offers, books appointments directly into the field service calendar, and upsells maintenance agreements — all without human staff involvement.
from callsphere import VoiceAgent, HVACConnector, CampaignManager
# Connect to HVAC service management
hvac = HVACConnector(
fsm="servicetitan",
api_key="st_key_xxxx",
calendar_lookahead_weeks=8
)
# Define the seasonal maintenance agent
maintenance_agent = VoiceAgent(
name="HVAC Maintenance Campaign Agent",
voice="lisa", # friendly, upbeat female voice
language="en-US",
system_prompt="""You are a friendly customer care representative
for {company_name}, an HVAC company. You are calling past
customers to offer seasonal maintenance service.
Your approach:
1. Greet warmly: "Hi {customer_name}, this is Lisa calling from
{company_name}. How are you today?"
2. Reference their history: "I see we last serviced your
{system_type} at {address} back in {last_service_date}."
3. Offer the seasonal service:
"We are scheduling {season} maintenance right now, and I
wanted to make sure you were taken care of before the
{peak_season} rush. A tune-up includes [service details]
and runs ${price}."
4. Handle objections:
- "I did it myself" → "That is great that you stay on top
of it! Our technicians also check refrigerant levels and
electrical connections that require specialized equipment."
- "Too expensive" → "We have a maintenance agreement option
that covers both seasonal visits for ${agreement_price}/year,
which saves you ${savings} and includes priority scheduling."
- "Not right now" → "No problem! When would be a better time?
I can set a reminder for you."
5. Book directly into the calendar if they agree
6. Offer the maintenance agreement for ongoing service
Be conversational, not pushy. If they are not interested,
thank them and move on graciously.""",
tools=[
"get_customer_history",
"check_calendar_availability",
"book_appointment",
"offer_maintenance_agreement",
"send_confirmation_sms",
"schedule_callback",
"update_customer_record"
]
)
@maintenance_agent.tool("check_calendar_availability")
async def check_calendar_availability(
customer_address: str,
preferred_date: str = None,
preferred_time_block: str = None # morning, afternoon, evening
):
"""Find optimal appointment slots based on route efficiency."""
# Get the customer's service zone
zone = await hvac.get_service_zone(customer_address)
# Find slots that optimize technician routing
available_slots = await hvac.get_optimized_slots(
zone=zone,
service_type="seasonal_maintenance",
preferred_date=preferred_date,
preferred_time=preferred_time_block,
optimize_for="route_density", # cluster nearby appointments
lookahead_weeks=6,
limit=5
)
return {
"slots": [
{
"date": slot.date,
"time_window": slot.time_window,
"technician": slot.assigned_tech,
"route_bonus": slot.route_efficiency_score
}
for slot in available_slots
],
"note": "Slots are optimized for route efficiency to "
"minimize drive time and reduce your wait window."
}
@maintenance_agent.tool("offer_maintenance_agreement")
async def offer_maintenance_agreement(
customer_id: str,
system_type: str
):
"""Present maintenance agreement options."""
customer = await hvac.get_customer(customer_id)
system_age = await hvac.get_system_age(customer_id)
# Customize agreement based on system age
if system_age and system_age > 10:
agreement_pitch = (
f"Since your {system_type} is over {system_age} years old, "
f"a maintenance agreement is especially valuable. Regular "
f"maintenance can extend the life of your system by 3-5 years "
f"and catch small problems before they become expensive repairs."
)
else:
agreement_pitch = (
f"A maintenance agreement covers both your spring and fall "
f"tune-ups for a single annual price, plus you get priority "
f"scheduling during peak season and 15% off any repairs."
)
agreements = [
{
"name": "Essential Plan",
"price": 189,
"includes": ["2 seasonal tune-ups", "Priority scheduling",
"10% repair discount", "Filter delivery"],
"savings_vs_individual": 49
},
{
"name": "Premium Plan",
"price": 299,
"includes": ["2 seasonal tune-ups", "Priority scheduling",
"15% repair discount", "Filter delivery",
"Indoor air quality check",
"Thermostat calibration",
"No overtime charges"],
"savings_vs_individual": 119
}
]
return {
"pitch": agreement_pitch,
"agreements": agreements,
"system_age": system_age
}
# Build campaign segments
customers = await hvac.get_customer_database(
has_phone=True,
exclude_active_agreement=True, # don't call existing members
exclude_do_not_call=True
)
# Segment by priority
segments = {
"high_priority": [
c for c in customers
if c.last_service_date and
(datetime.now() - c.last_service_date).days > 365 and
c.system_age and c.system_age > 8
],
"medium_priority": [
c for c in customers
if c.last_service_date and
(datetime.now() - c.last_service_date).days > 180
],
"agreement_upsell": [
c for c in customers
if c.total_service_calls > 2 and
not c.has_maintenance_agreement
]
}
# Launch the spring AC maintenance campaign
for segment_name, segment_customers in segments.items():
await maintenance_agent.launch_campaign(
customers=segment_customers,
segment=segment_name,
calls_per_hour=100,
calling_hours={"start": "09:00", "end": "19:00"},
calling_days=["monday", "tuesday", "wednesday",
"thursday", "saturday"],
timezone_aware=True,
retry_on_no_answer=True,
max_retries=2,
retry_delay_hours=48,
campaign_name="Spring AC Maintenance 2026"
)
| Metric | Without AI Campaign | With AI Campaign | Change |
|---|---|---|---|
| Shoulder-season utilization | 38% | 81% | +113% |
| Maintenance appointments/month | 45 | 280 | +522% |
| Maintenance agreement sign-ups | 12/month | 85/month | +608% |
| Agreement annual revenue | $27K | $192K | +611% |
| Off-peak monthly revenue | $52K | $134K | +158% |
| Customer contact rate (database) | 3% | 62% | +1,967% |
| Cost per appointment booked | $35 | $4.50 | -87% |
| Equipment replacement leads | 8/month | 34/month | +325% |
Metrics from an HVAC company (12 technicians, 5,200 customer database) deploying CallSphere's seasonal campaign agent over one spring cycle.
Week 1: Export and clean your customer database from ServiceTitan, Housecall Pro, or your FSM platform. Validate phone numbers and tag customers by system type (AC, furnace, heat pump), last service date, and system age. Connect CallSphere to your FSM calendar for real-time availability.
Week 2: Configure seasonal scripts (spring = AC focus, fall = furnace focus). Set up maintenance agreement offerings and pricing. Define route-optimized scheduling zones. Test with 100 simulated calls using real customer profiles.
Week 3: Launch the campaign 6-8 weeks before peak season. Start with the highest-priority segment (customers with aging systems and lapsed maintenance). Monitor booking rates and agreement conversion daily.
Still reading? Stop comparing — try CallSphere live.
See the home services AI agent handle a real call — complete, industry-specific, and live in your browser. No signup.
Week 4-6: Expand to remaining segments. The AI agent fills the schedule progressively, creating dense appointment clusters that minimize technician drive time. CallSphere's route optimization typically reduces drive time by 25-35% compared to manually scheduled appointments.
An HVAC company in the Sun Belt region deployed CallSphere's seasonal campaign agent for their spring 2026 AC maintenance push:
The operations manager summarized: "We used to dread April and May. Techs were sitting around, and I was worried about making payroll. Now those months are almost as busy as July, and the revenue from maintenance agreements alone covers our off-peak overhead."
Start 6-8 weeks before your peak season begins. For AC maintenance, launch in mid-March to early April. For furnace maintenance, launch in mid-September to early October. This gives enough lead time to fill the schedule progressively and ensures customers are thinking about their systems before they actually need them. CallSphere can schedule campaigns to auto-launch based on date ranges.
Data from CallSphere's HVAC campaigns shows the highest contact and conversion rates on Saturday mornings (9am-12pm) and weekday evenings (5pm-7pm). Midday weekday calls (11am-2pm) have surprisingly good contact rates with retirees and work-from-home customers. The AI agent automatically adjusts calling patterns based on contact rate data for your specific customer base.
The agent does not know about past complaints unless you flag those customers in the database. Best practice is to exclude customers with unresolved complaints from automated campaigns and have a human manager reach out to those customers separately. For customers who mention a past issue during the call, the agent acknowledges the concern, apologizes, and offers to have a manager call them back to make it right.
The agent does not close equipment sales (which typically require an in-home assessment), but it excels at identifying replacement opportunities. When a customer mentions an aging system, unusual noises, rising energy bills, or frequent repairs, the agent flags the lead and offers to schedule a free in-home assessment. These warm leads convert to equipment sales at 35-45%, compared to 8-12% for cold leads.
Written by
Sagar Shankaran· Founder, CallSphere
Sagar Shankaran is the founder of CallSphere, where he builds production AI voice and chat agents deployed across healthcare, hospitality, real estate, and home services. He writes about agentic AI, LLM engineering, and shipping voice agents that handle real calls in production.
See how AI voice agents work for your industry. Live demo available -- no signup required.
A how-to for Colombian education and tutoring SMBs to answer parents and students instantly, book trial classes 24/7 in Spanish and English, and grow enrollment with a CallSphere AI agent.
Tbilisi professional-services firms serving relocating founders and IT companies use CallSphere AI voice and chat agents to answer enquiries 24/7 in English, Georgian and Russian and book consultations.
A practical how-to for Palau eco-resorts and dive operators on capturing every high-value, multilingual enquiry with a CallSphere AI voice and chat agent, while honouring Palau’s marine-conservation commitments.
How salons, spas and wellness SMBs across the UAE, Saudi Arabia and Qatar use CallSphere AI voice and chat agents to capture every booking 24/7 in Arabic, English and expat languages, and cut no-shows.
How estate agents and property managers in Luxembourg City and across the Grand Duchy use CallSphere to capture multilingual viewing and enquiry calls 24/7, GDPR compliant.
A practical guide for Senegalese logistics, freight, legal and consulting SMBs in Dakar and Thiès to capture every enquiry 24/7 with a CallSphere AI voice and chat agent in French, Wolof and English.
© 2026 CallSphere LLC. All rights reserved.
Made within New York
Watch how CallSphere handles real customer calls, schedules appointments, and processes payments — live.
Try Live DemoBook a DemoCalculate Your ROI