Most home service companies are good at the visit. A tech or estimator walks the job, measures, explains the options, and leaves a quote. Then the quote sits. The office is busy with today's calls, the estimator is already at the next house, and nobody is sure whose job it is to check back.
If you run an HVAC, plumbing, roofing, pool, pest, or remodeling company in Jacksonville, Orlando, or anywhere in Florida, open estimates are one of the easiest places to apply automation. The work is repetitive, the timing matters, and the result is simple to measure: did the customer say yes, no, or nothing?
This guide covers how to set up estimate follow-up that is consistent, respectful, and secure, where AI actually helps, and how to tell whether it's working.
Why open estimates slip through the cracks
Follow-up usually fails for ordinary reasons, not because anyone doesn't care:
- No clear owner. The estimator thinks the office will call. The office thinks the estimator will.
- No list. Open quotes live in a field service app, a spreadsheet, a stack of paper, and someone's text messages.
- Awkward timing. Calling too soon feels pushy. Waiting too long means the customer already hired someone else or forgot the details.
- No record of what happened. Without notes, nobody knows whether the customer said "after the holidays" or "too expensive."
Automation doesn't fix a weak offer or a bad visit. It fixes the gap between a good visit and a decision.
What a good follow-up sequence looks like
Start simple. A sequence for most residential estimates can be three or four touches over two to three weeks, then a clean stop.
Step 1: Send the estimate the same day
The first follow-up is really the estimate itself. Send it while the visit is fresh, with the options laid out plainly and a direct way to ask questions. If your field service software already does this, make sure it actually goes out every time.
Step 2: A short check-in a few days later
A brief message asking whether the customer has questions works better than "just following up." Mention the specific job so it doesn't read like a mass blast.
A sample text might look like this (hypothetical example):
"Hi Dana, this is the office at [Your Company]. Checking in on the AC replacement options from Tuesday. Happy to answer any questions or adjust the quote. Reply STOP to opt out."
Step 3: A helpful follow-up a week or so later
Give the customer something useful instead of pressure: answers to a common question about the job type, a reminder about scheduling lead times, or an offer to walk through the options by phone.
Step 4: A polite close
After the last touch, ask whether they'd like you to keep the quote open or close it out. Many people reply to this one because it is easy to answer. Then stop. Endless reminders hurt your reputation more than they help.
Who handles the replies
Any reply, question, or "yes" should go to a person right away. The automation's job is to start conversations and keep the list current, not to negotiate or answer technical questions on its own.
Where AI helps, and where it shouldn't
AI is useful here, but only in specific spots.
Good uses:
- Drafting personalized messages from the estimate details, so each check-in mentions the actual job instead of a generic line. A person can review drafts before they go out, at least at first.
- Sorting replies. AI can tag responses as "ready to book," "has a question," "price concern," "not now," or "no," and route them to the right person.
- Summarizing notes. After a call, AI can turn rough notes into a clean summary on the job record, so the next person knows what was said.
Keep a person in charge of:
- Price changes, discounts, and financing questions.
- Anything about safety, code requirements, or warranties.
- Complaints or upset customers.
- Deciding when to stop following up.
A simple rule: AI can draft and sort, but a human decides anything that changes the price, the promise, or the relationship.
Texting rules and consent
Text messages get read, which is why they're useful, and also why they come with rules. Before you automate texts:
- Get consent. Make it clear on your booking form or estimate that the customer may get texts about their estimate, and keep a record of that consent.
- Include an opt-out in every message. "Reply STOP to opt out" should be in each text, and opt-outs must be honored right away, across every system that sends messages.
- Keep it about the job. Estimate follow-up is about a service the customer asked for. Don't quietly turn the list into a marketing list.
- Send at reasonable hours. Set your system to send only during normal daytime hours in the customer's time zone.
Texting and telemarketing rules change and vary by situation, so check with your attorney or your messaging provider about current requirements before launching.
Keep customer data secure
Estimate records often include home addresses, phone numbers, photos of the inside of someone's home, and sometimes gate or lockbox codes. Treat that data with care:
- Limit what goes into AI tools. The message drafter needs the customer's first name, the job type, and the options. It doesn't need gate codes or payment details.
- Use business accounts with proper data controls, not personal chatbot accounts, for anything involving customer information.
- Control access. Only the people who handle estimates should see the full pipeline, and former employees should lose access the day they leave.
- Keep an audit trail of what was sent, when, and by which system, so you can answer questions if a customer asks.
How to measure whether it's working
The point of automation is a better result, not just less effort. Measure before and after.
Set a baseline first
Before turning anything on, pull your last couple of months of estimates and write down:
- How many estimates you sent.
- How many turned into booked jobs.
- How long it typically took from estimate to decision.
- How many open estimates never got any follow-up at all.
If you don't have clean data, track the next two weeks by hand. A rough baseline beats none.
Track the same numbers after launch
Run the new process for a full month or two, then compare:
- Close rate on estimates, overall and by job type.
- Time to decision.
- Share of estimates with a recorded outcome, meaning yes, no, or not now instead of silence.
- Opt-outs and complaints. If these climb, your messages are too frequent or too pushy.
- Office time spent chasing quotes.
If close rates don't improve but every estimate now has a clear outcome, that's still useful: you'll know why jobs are lost and can fix pricing, scheduling, or the visit itself.
Common mistakes to avoid
- Automating everything at once. Start with one estimate type, such as AC replacements or repipes, and expand after you see results.
- Sending the same message to everyone. Generic follow-ups get ignored.
- Letting AI reply without review. Especially early on, keep a person approving messages.
- Forgetting to stop. Close the loop after the last touch.
- Not connecting it to scheduling. When a customer says yes, booking should be one step, not a new round of phone tag.
FAQ
Do I need new software to do this? Often not. Many field service platforms include estimate follow-up features that go unused. Start by checking what you already pay for before adding anything.
How many follow-ups is too many? For most residential jobs, three or four touches over a few weeks, followed by a clear close, is a reasonable starting point. Watch opt-outs and replies and adjust.
Should follow-up be by text, email, or phone? Usually a mix. Texts get noticed, emails hold the details, and a phone call works best for larger jobs or customers with questions. Let customers tell you how they prefer to be reached.
Is this only for big companies? No. A two-truck operation with a part-time office manager often benefits the most, because there's no spare time to chase quotes by hand.
Want help setting it up?
If you'd like help mapping your estimate process, setting up a follow-up sequence that respects your customers, and measuring whether it wins more jobs, we can help you pick one workflow and measure it. Take our free assessment at rsoftware.net/assessment or email us at hello@rsoftware.net.
For more practical guides on secure AI, automation, and security for local businesses, follow R Software & Consulting on LinkedIn.
Written with help from our AI assistant.
Want a rough estimate of what automation could save each year?
A 30-minute call, at no cost.
Get my free savings snapshot →Follow R Software & Consulting on LinkedIn for practical, secure AI ideas for local small businesses. Follow us on LinkedIn (opens in a new tab)