Recovering missed HVAC calls

The company was generating inbound calls through Google Business Profile, Google Ads, and referrals, but missed calls were going unanswered, particularly during busy periods and evenings.

Staff would often call back hours later, by which point the homeowner had already contacted another HVAC company. The owner knew leads were coming in but had no reliable way to distinguish between lead volume and actual booked jobs.

Marketing was not the constraint. The leak sat between the ring and the booking: a warm inbound call became a competitive bid the moment the callback took hours.

Because nothing was logged against the job that eventually got booked, the owner was watching the wrong number. Lead volume can rise while booked work stays flat, and the reports will still look healthy.

Respond in minutes, not hours, and only automate the intake. The job was to catch the call, qualify it, and get it on the board.

Deliberately parked: an AI that diagnoses equipment or gives technical advice. That would have been a liability change. Speeding up the response is a workflow change; pretending to be a technician would have been a risk change.

Built a lightweight AI-assisted intake and follow-up system. When a call was missed, it automatically sent a text acknowledging the request and collected service address, HVAC issue, urgency, and preferred appointment time.

An AI phone receptionist covered periods when the office could not answer, limited to intake, qualification, scheduling, and escalation. Every interaction was logged against the eventual booked job, so the owner could see which marketing sources actually produced paying customers.

A lead you answer too late is not a lead. It is a referral to the competitor who picked up.

Cost per lead is a vanity metric when the leak is in the callback. Cost per booked job is the number that can change a spend decision.