Automating HVAC Quote Follow-Up (No Robocall Feel)

The Quote Goes Out, Then Everything Goes Quiet

You send a homeowner pricing on a furnace or AC install. They said they wanted it "this week." Three days pass and nothing comes back. Someone in the office calls, gets voicemail, leaves a message asking them to call back at their earliest convenience. Two days after that, a follow-up email goes out with the PDF attached again, subject line something like "Following Up." Still nothing. By the time anyone tries a third time, the homeowner has already booked with whichever company called them back first.

Across the HVAC shops we've worked with, a large share of quotes never convert, and it usually isn't because the number was wrong. Homeowners shopping anything above a basic repair typically collect two or three bids before deciding, and the contractor who stays visible through that window closes more of them than the cheapest one does. Price matters less than whoever shows up again at the right moment with the right tone.

Most HVAC businesses already do some kind of follow-up, so that part rarely goes missing. What actually goes wrong is the follow-up itself, since it tends to sound like every other company's follow-up, like a script somebody wrote without ever having to convince a homeowner to spend six grand on a new furnace.

Why Voicemail and Templated Email Don't Work Anymore

Picture the voicemail: "Hi, this message is for John, this is Sarah calling from ABC Heating and Cooling to follow up regarding the estimate we provided on the 14th, please give our office a call back at your earliest convenience." That's a script built for a phone system rather than a person on the other end, and homeowners have gotten very good at screening calls from numbers they don't recognize. Most of these voicemails get deleted without a listen past the first five seconds.

Email has the same issue in a different form. "Just following up on your quote, let us know if you have any questions!" reads exactly like every drip campaign that's ever landed in someone's inbox, and it usually sits there unopened, because there's nothing in the subject line that suggests a real person wrote it. Quotes for HVAC work also tend to get sent as PDFs, which means the follow-up email is asking someone to reopen an attachment they already glanced at once and set aside.

None of this is really the fault of whoever's sending it. Most shops don't have time to write a fresh personal message to every quote that goes quiet, so they fall back on the same handful of templates every quote gets, and that habit is quietly doing most of the damage to the response rate.

What a Better Sequence Actually Looks Like

What we build for HVAC clients instead is a short SMS and WhatsApp sequence that runs off Twilio, orchestrated in Make, aiming to sound like the technician who did the estimate is texting the homeowner directly, because that's basically what's happening.

Here's the shape of it, step by step:

Trigger: the quote gets marked "Sent" in HubSpot, or if the shop is running things off a spreadsheet (a lot of them are, and that's fine), a row gets a status of "Quote Sent" with a timestamp in Google Sheets.

Same-day touch, a few hours later: a short text goes out through Twilio. Something like "Hi Mike, it's Dave from Coastal Air. Sent over the pricing on the furnace earlier, let me know if anything doesn't make sense or if you want to talk through options." No company logo or greeting-card formatting, just a text from a name the homeowner already met.

Day 3, if there's no reply logged: a second message goes out, this time over WhatsApp if the customer opted into it on the quote form, or SMS if not. Slightly softer, maybe references financing or timing: "Hey Mike, checking if the numbers worked for you or if timing's the holdup. Happy to jump on a quick call if that's easier than texting back and forth."

Day 7, final check-in: short, low-pressure, with an easy out. "Hi Mike, last note from us on this one for now. If plans have changed no worries at all, just let us know and we'll circle back down the road instead."

Reply handling: any incoming reply gets routed back into HubSpot (or the sheet) and pings the rep on Slack or forwards straight to their phone, because the whole point is getting a human into the actual conversation. Automation only needs to land the first reply, closing still happens person to person after that.

Make sits in the middle of all this, watching for the status change, waiting the set interval between touches, checking the quote hasn't already moved to Won, Lost, or Replied before sending the next message, and firing the Twilio API call for SMS or the WhatsApp Business API for the WhatsApp leg. Shops already running n8n can build the same logic there instead, the tool matters less than the sequence and the tone. For anyone without a CRM at all, this runs perfectly fine off a Google Sheet with a date column and a status column. A lot of HVAC shops are already tracking quotes that way, CRM or not. It's the same watch-and-wait pattern behind getting a job dispatched to the right tech the moment it's booked, just applied earlier in the process, before the job is even won.

The Objection We Always Get: "Won't This Feel Like We're Chasing Them?"

Fair question, and worth taking seriously given this is a big-ticket home purchase rather than a $40 impulse buy.

The short answer is that three short, spaced-out messages over a week read completely differently than a daily nudge or a scripted robocall. What actually annoys people is a call center reading from a screen, or a message that clearly went to five hundred numbers at once. A text from "Dave" that references the actual job, sent a handful of times over seven days with a clear low-pressure close on the last one, tends to land as attentive instead of pushy, and that mostly comes down to the volume staying low while the tone does the actual work.

Worth also being upfront that this needs consent baked into the quote request form itself, a simple checkbox or line of copy saying the homeowner agrees to be contacted by text about their quote. Most places require that for SMS outreach anyway, and it's a five-minute fix on the form if it isn't there already.

What This Actually Recovers

Say a shop sends out 40 quotes a month and converts roughly a quarter of them with their current follow-up. Pulling even an extra 4 or 5 of those quotes back into the funnel each month, on install jobs averaging somewhere between $6,000 and $9,000, works out to real money fast, often in the $20,000 to $40,000 range depending on average ticket size and how many of those recovered quotes are installs versus smaller repair jobs. That kind of recovery doesn't need a single extra quote going out the door, just the same leads followed up with in a way that actually gets a reply. Money you would've otherwise just left sitting in a "no response" column somewhere.

There's also a quieter benefit that doesn't show up in the revenue math directly: the office stops spending an hour a day on manual callbacks and voicemail drops that mostly go nowhere anyway, and that time goes back into estimating or scheduling instead.

The Tools This Actually Runs On

We don't write custom code for this kind of build unless there's genuinely no off-the-shelf way to do it, which for a sequence like this there almost never is. It runs on tools most shops either already have or can get running the same week: Twilio handles the SMS and WhatsApp layer (worth a look at what this kind of message volume actually costs you on Twilio before committing to a sequence), Make handles the logic connecting it to HubSpot or a Google Sheet, and whatever CRM or spreadsheet is already tracking quotes stays right where it is, since getting the CRM side dialed in for an HVAC shop usually matters more than which platform it happens to be.

If your quotes are going quiet and the follow-up you've got feels more like a script than a conversation, we'll take a look at your current process for free. Book a free 60-minute call and we'll map out where quotes are dying, what a sequence like this would look like for your specific jobs, and what it'd cost to get it running.