Agents in the Trades
Communication

The tune-up list that texts itself before the season turns

It is the last slow week of September and the office manager at a mid-size HVAC shop is staring at a spreadsheet of maintenance customers due for tune-ups before heating season — and at a dispatch board with open slots next week.

Source: Contracting Business, "AI Finds Its Place in the HVAC Business" — https://www.contractingbusiness.com/contracting-business-success/article/55396975/ai-finds-its-place-in-the-hvac-business — the article describes Jolly Heating Air Plumbing and Electric, serving Cincinnati and Northern Kentucky, using Probook AI to send outbound text messages for maintenance reminders, scheduling and promotions, alongside answering chats and calls, helping with booking and dispatch, and generating KPI reporting, with staff stressing that the tool has to be trained and coached like a new employee. Adapted here for a small shop; no vendor or shop results are claimed.

The problem

Every shop has the same list: customers who signed a maintenance plan or came in for a tune-up last fall and are due again. The work is already sold in principle. What is missing is the hours to chase it — calls nobody answers, postcards nobody reads, and a reminder task that always loses to today’s breakdowns.

So the shoulder season stays soft for no good reason. Techs sit under-booked before the first cold snap, then get buried once it hits. The customers were there the whole time, in the shop’s own records, waiting for a nudge in a form they would actually answer.

Get one workflow like this, every day.

The free Agents in the Trades newsletter — each workflow broken down: what it does, what it costs, and where it breaks.

How the agent workflow works

  1. The due list builds itself from records you already have. The agent pulls customers by last tune-up date, plan status, and equipment, sorted by who is due first — not by whoever is on a sticky note.
  2. Each customer gets a text that sounds like the shop, not a blast. It names the service due and offers real open windows, in the shop’s wording, at hours a homeowner might reply.
  3. Replies continue as a conversation, not a campaign. The agent keeps the thread going, answers the routine parts, and writes the outcome back to the customer record.
  4. Booking lands on the real board. A confirmed tune-up goes onto dispatch with the right duration and job type, matched where possible to a tech who knows that system.
  5. Promotions use the same discipline. A seasonal offer only goes to customers it fits, by equipment and history, not to every number the shop has ever collected.
  6. The office sees what the outreach produced. Booked, replied, declined, and unreachable come back as a simple roll-up — the KPI side of the same workflow — so the next round starts from results.

What you need to set it up

Where it breaks

A text to the wrong person is worse than no text. Old numbers get reassigned, tenants move, and a reminder about a furnace at an address someone left two years ago reads as careless. List hygiene and a human review of the first batch are not optional.

Outbound fills the board in reply order, not dispatch order. A successful round can stack tune-ups across town on the same morning and leave no room for the breakdown that calls in at eight. Cap each day’s slots and hold emergency capacity back before the first message goes out.

Promotions train customers to wait for a discount. If every quiet week triggers an offer, full-price maintenance gets harder to sell, and a reminder thread becomes a negotiation the agent cannot run. Keep promotions rare and targeted, and route pricing pushback to a person.

The bottom line

Start with one season change and one clean segment — plan members due in the next thirty days — and review every thread for a week. The win is not more texts. It is turning a list the shop already owns into booked, well-routed tune-ups before the weather does the marketing.

Related use cases

← All use cases

One working agent workflow, every day.

Real AI agent use cases from solar, construction, and field operations — what they do, what they cost, and where they break. No hype, no theory.