Methodology · v1.0
How we recover an HVAC shop's old estimates
Second Swing works a contractor's unclosed estimates in four steps: confirm the shop had consent to contact each person, sort every estimate by what stopped the sale, follow up in the shop's name during legal hours, and hold back a fifth of the list at random to prove the callbacks came from the work and not from luck.
Most follow-up services can't tell you whether they did anything. They report what closed and let you assume they caused it. The method below is built around one idea: separate the jobs we caused from the jobs that would have closed anyway. Everything else serves that.
The four steps
- Confirm consentWe only contact people who already asked the shop for an estimate. That existing relationship is the legal basis for the outreach. The shop exports its unclosed estimates from the last 180 days and confirms these are their own prospects. No purchased lists, no cold numbers.
- Sort by what stopped the saleEvery estimate gets tagged by objection: price, financing, timing, still shopping, spouse or partner decision, repair versus replace, or already bought elsewhere. The reason it stalled decides the message. A financing stall and a "getting other quotes" stall are different conversations.
- Follow up in the shop's name, within legal hoursThe message goes out under the shop's own identity, not ours. Nothing sends before 8am or after 9pm in the customer's local time, every message carries an opt-out, and anyone who replies STOP is removed for good. The shop approves the wording before anything goes out.
- Hold back a fifth to measureTwenty percent of the list is set aside at random and never contacted. At day thirty we compare the group we worked against the group we left alone. The difference in closed jobs is the recovery. If there's no difference, the shop finds out from us.
Why the holdout group matters
Some of your old estimates were going to close no matter what. A homeowner circles back, a season turns, the old unit finally dies. If a follow-up service counts those as wins, the number is fiction. The holdout group is the control. It's the same list, same period, left untouched, so the only difference between the two groups is us. That's the difference between a report and a result.
What we do, what you do
We sort the list, write the follow-up, run it compliantly, and hand back booked callbacks with the original quote and one line on what held the person up. You call the warm callbacks back, within about four hours, and close them. Booking is ours. Closing is yours. The shops that call back fast get the most out of this. The ones that let a callback sit two days lose it.
Compliance, plainly
Texting past customers is governed by the TCPA and by carrier rules for business messaging (A2P 10DLC). We handle registration, quiet hours, and opt-outs so the shop stays clean. The short version: this is a warm follow-up to people who raised their hand, sent on the shop's own registered number, at reasonable hours, with a working opt-out. That is the line between a welcome reminder and a violation.
Version history
v1.0, July 28, 2026. Four-step method with a 20% random holdout as the standard proof of lift. Los Angeles HVAC replacement and major repair. Future versions will note any change to the holdout share, the objection set, or the measurement window.
Work my backlog