How to Bill HVAC Maintenance Plans Automatically (Without Chasing Cards)

A maintenance agreement is the closest thing the trades have to recurring software revenue. The customer pays a fixed amount every month, you show up twice a year for a tune-up, and in exchange you get first call when the system fails in August. Shops that run a healthy plan book quote more replacements, carry fewer idle technician hours in the shoulder seasons, and sell at a higher multiple when the owner retires.

The economics only work if the money actually arrives. A 200-plan shop billing $25 a month is collecting $5,000 a month, and in most shops a meaningful slice of that arrives as checks in an envelope, cards read over the phone, or an annual invoice someone forgets to send. This article walks through pricing a plan, putting it on automatic monthly billing, and handling the two things that quietly erode a plan book: expired cards and cancellations you never see coming.

Price the plan monthly, not annually

Most maintenance agreements were originally sold as an annual amount, $180 or $300 for the year, invoiced once. Annual billing looks simpler on paper and is worse in practice. It creates one large decision every twelve months, which is exactly when a customer reconsiders whether they need you. It also concentrates your collection risk: one failed payment is a whole year of revenue, not a month.

Monthly pricing between $15 and $35 covers most residential markets. The band matters less than the structure. A plan at $22 a month reads as a utility line rather than a purchase, and the customer who is annoyed in month seven usually stays because cancelling requires a phone call about $22, not $264.

• Set the monthly amount from your actual visit cost, not a competitor’s flyer. Two tune-ups, a filter, and a priority-dispatch promise have a real cost per plan.
• Decide what the plan includes in writing: number of visits, the discount on repairs, whether after-hours dispatch is covered.
• Charge the same day every month. Customers reconcile their statements; predictability reduces disputes.
• Keep a small annual option for the customers who insist, priced at eleven months of the monthly rate rather than a deep discount.

Get the plan on automatic billing at the point of sale

The single biggest determinant of whether a plan book collects cleanly is when you capture the payment method. A card captured in the customer’s living room, while the technician is packing up after a repair, converts and stays. A card requested by email three days later does neither.

That is a practical argument for billing from the phone the technician already carries. In the Paydigo merchant app, a technician can create the plan, take the first payment on the spot with Tap to Pay, and put the customer on a monthly cycle before leaving the driveway. The card is stored as a token, not on a clipboard, and the customer gets a receipt by text before the truck pulls away.

If you sell plans over the phone, use a payment link instead of reading numbers aloud. The customer taps the link, enters their own card, and the plan starts from their entry. Nobody in your office ever handles the digits, which is both a compliance improvement and a real reduction in typos.

Expect failed cards and plan for them

Every subscription book leaks. Cards expire, get reissued after a breach, or bounce because the customer moved money around. A plan book with no recovery process loses a predictable percentage of its revenue every year, and the loss compounds because a cancelled plan is also a lost replacement lead.

Three mechanisms cover most of it. Automatic card updating catches reissued cards without anyone calling the customer. Retry logic re-attempts a declined charge on a schedule instead of writing it off. And a customer-facing update link lets the homeowner fix their own card in thirty seconds, which is how most failures actually get resolved, because the customer wants the plan.

What you should not do is treat a failed charge as a collections event. The customer whose card was reissued is not a deadbeat; they are a plan member with a stale token. Automate the recovery and reserve human follow-up for the accounts that fail twice.

Know what the plan book costs you to run

Two costs matter: the transaction cost of collecting each payment, and the platform cost of running the subscriptions themselves.

On transaction cost, a plan payment is a card-not-present charge, the card is stored, not tapped, so it prices at the higher of the two card rates. With Paydigo that is 2.90% + 30¢. On a $25 plan payment that is about $1.03, or roughly four percent of the payment. Bank payments are the lever worth knowing about: ACH runs 1% capped at $10, so on larger commercial agreements the cap does real work.

On platform cost, running subscriptions is the one part of Paydigo that carries a monthly fee: the subscriptions platform is $49 a month, with a 30-day free trial. Invoices, payment links, QR codes and Tap to Pay are free. For a 200-plan shop, $49 against $5,000 collected is under one percent of the book, but it is a real line, and the trial exists so you can prove the collection improvement before you pay for it.

Sell the plan the way the customer experiences it

The plans that survive renewal season are the ones where the customer can name the benefit. "Two tune-ups a year and you go to the front of the line in a heat wave" is a benefit. "Priority membership tier" is not. Write the plan description in the language the homeowner would use to explain it to a neighbor, and put that language on the receipt they get every month. A recurring charge with a vague label is a common cause of a friendly-fraud dispute.

Label the charge with your business name and the plan name, both of which show on the customer’s statement. When a customer does dispute a plan payment, dispute alerts give you the chance to resolve it as a refund before it becomes a filed chargeback, which saves the $20 chargeback fee and, more importantly, keeps your dispute ratio clean.

A workable rollout for an existing plan book

If you already have a hundred agreements on paper, do not try to migrate them all in a week. Run the conversion through your normal service calendar.

• Week one: put every new plan sold on monthly automatic billing. No exceptions, no paper.
• Weeks two through eight: as each existing plan comes up for its scheduled visit, have the technician convert it in the app during the visit.
• Month three: send a payment link to the remaining holdouts with a short note explaining that the plan is moving to monthly and the amount is smaller each time.
• Month four: reconcile. Count active plans, monthly collected, and failed-payment recoveries. This is the number that tells you whether the plan book is real.

Shops that do this stop discovering in November that forty agreements quietly lapsed in June. The plan book becomes a number you can read on a dashboard, which is the first step to managing it like the asset it is.

Handle cancellations as data, not as loss

A plan book with no cancellations is a plan book nobody is auditing. What matters is knowing why each one left and how quickly you heard about it. Cancellations that arrive as a phone call are useful; cancellations you discover six months later, when a technician notices the tune-up was never scheduled, are the expensive kind.

Set one rule: every cancellation is recorded the day it happens, with a reason from a short fixed list, moved, sold the house, replaced the system, price, service complaint, no reason given. Six categories is enough. Within two quarters you will know whether you have a pricing problem, a service problem, or ordinary churn from people leaving the area, and those three problems have completely different fixes. Owners who skip this step tend to respond to any cancellation by discounting the plan, which is the one response that hurts a healthy book.

Two numbers to read every month

Plan books are managed with fewer metrics than people expect. Active plan count tells you whether selling is working. Monthly recurring amount collected, not billed, collected, tells you whether the billing is working. The gap between the two is your leakage, and it should be small and shrinking.

If billed and collected diverge by more than a few percent, the cause is almost always stored cards rather than unwilling customers. Work the failed-payment list weekly, in one sitting, and the gap closes. If active plans are flat while service calls are growing, the problem is that technicians are not offering plans on repair visits, a coaching issue, not a billing one.

What to take away

Maintenance agreements are worth automating because the failure modes are boring and fixable. Price monthly, capture the card while the technician is still on site, let the platform handle expired cards, know your per-payment and platform costs, and label the charge so the customer recognizes it. Do those five things and a plan book behaves like recurring revenue instead of a filing cabinet.