How to sell HVACin the shoulder season

The trough is not a demand problem. It is an offer problem, and the offers that fix it are not the ones you run in July.

Every contractor in this trade knows the shape of their own year without needing a chart. The phone rings itself off the hook in July, dies in October, wakes up briefly when the first cold snap arrives, and dies again in April. Payroll does not follow that pattern. Neither does rent, or the truck note, or the ad account you switched off in week three of the quiet spell.

What follows is the version of the problem worth solving, and the four plays that solve it.

The shape you are working against

Here is the curve, averaged across Southern California. Your own market will differ in depth and timing, sometimes drastically, and that matters more than most contractors realise. More on that further down.

The instinctive responses to that dip all make it worse. Switching campaigns off surrenders the conversion data the platform spent the busy season learning, so the following June you start cold and pay a premium to relearn it. Discounting tune-ups fills the calendar with your lowest-value job at the moment your techs are cheapest to deploy, which trains the market to wait for a sale. And cutting crew in October means recruiting in May, in a labour market where everyone else is doing the same thing.

Play one: sell plans in the trough, not the peak

A maintenance agreement is not a discount you give away. It is the only product in your catalogue whose value compounds, and the shoulder months are the correct time to sell it because your technicians have the time to do the visits properly and your customers are not in distress.

The economics are worth seeing laid out, because most owners price plans as though the fee were the point. It is not. The fee buys you access.

The last segment is the whole argument. A plan member who needs a system replaced does not open three tabs. You have been inside the house twice a year, you know the equipment, and the quote arrives as a continuation rather than a competitive bid. That difference in close rate is where the money is, and it is invisible if you judge plans on the fee alone.

It also changes what you can afford to pay for a customer. Run the max cost per lead calculator twice, once on a repair invoice and once on plan lifetime value, and the second number will usually justify bidding you had ruled out. That is the practical use of this article: the trough is affordable to attack precisely because you are buying a longer relationship, not a job.

Play two: pull replacement demand forward with financing

Nobody replaces a working system in April because they want to. They will replace one in April if the decision is framed as a monthly number and the alternative is a failure in July when you are three weeks out and charging peak rates.

That argument only lands if you make it before the failure, which means the shoulder campaign is aimed at a different audience than the peak campaign: homeowners with equipment at replacement age who are not currently in any distress. They are cheap to reach in a quiet month and they convert on a longer cycle that your follow-up has to survive.

Two things make it work. Put the monthly figure in the creative rather than the system price, and give the offer a reason to be now: install availability, a rebate deadline, a pre-season price hold. A reason that is true, because a manufactured deadline gets recognised.

Play three: work the database you already paid for

Your CRM is full of people who called once, took a quote and vanished, and every one of them was paid for at full price. That list sits untouched because working it is nobody's job when the phone is busy and everybody's last priority when it is not. The shoulder months are when it should be worked, systematically:

• Unsold replacement quotes from the last twenty-four months, by age of quote.
• Systems you have already inspected and flagged as end of life.
• Lapsed maintenance agreements, which are cheaper to reinstate than to originate.
• Customers with a repair history who have not been contacted in a year.

This is what automation is genuinely good at: the work is repetitive, the timing matters more than the wording, and nobody in your office will do it consistently through a busy August. Set the sequences once and let them run against the calendar rather than against somebody's memory.

Play four: buy a floor

Everything above smooths the curve. Contracted commercial work removes it. A service agreement on a small commercial portfolio is revenue that arrives in February whether or not anybody is uncomfortable, and it changes the shape of the business rather than the shape of the quarter.

It is slower to build than any campaign. Facilities managers buy on reliability and documentation, they are reached by outreach rather than advertising, and the cycle runs months, which is exactly why the shoulder season is when to do it. In denser markets with a light industrial corridor, this is less a supplement than a second business sitting unadvertised.

Your trough may not be where you think

The curve at the top of this page is an average, and averages are dangerous here. Before you build a calendar around it, check your own market, because across Southern California the differences are large enough to invert the advice.

• The Inland Empire runs the canonical version: a long, brutal summer peak and genuinely empty spring and autumn quarters. Everything above applies as written.
• The Coachella Valley inverts it entirely. Peak failure season lands when a third of the owners are out of town, and the quiet months for equipment are the busy months for population. The shoulder campaign there runs November to April.
• The Conejo Valley barely has an autumn trough at all. Santa Ana heat, smoke events and power shutoffs stack into September through December, which is precisely when competitors wind their spend down.
• In the South Bay the whole calendar reverses, because heating demand outweighs cooling. Summer is the shoulder there, and the winter auction thins out just as the phone gets busy.
• On the Ventura coast there is no meaningful seasonal shape in either direction, which sounds restful and is actually harder: no peak means no natural urgency to sell against, and the work has to come from contracts and equipment age instead.

Pull your own Trends data before you plan a budget calendar. Twenty minutes in a browser will tell you whether the advice above needs shifting by two months or turning upside down.

What to spend, and when

The counterintuitive part is that the trough is usually the cheapest attention of the year. Competitors follow the demand curve with their budgets, so cost per click falls in the months when almost nobody is bidding. In most Southern California markets we see a meaningful discount on shoulder-month clicks, though it is worth checking in your own geo rather than assuming.

So the calendar looks roughly like this. Hold a floor of spend through the trough on plan, financing and database offers. Ramp hard in the four to six weeks before the peak rather than during it, so the account has conversion data and momentum when the auction gets expensive. Then let the peak pay for itself on emergency and replacement intent, and use the season to sell the plans that carry the next trough.

One number worth tracking that almost nobody does: replacement close rate by month of quote. Quotes issued in a shoulder month tend to close at a different rate than the same quote issued in the middle of a heat wave, because the buyer has time to think and you have time to sell properly. If that gap is real in your data, it is an argument for moving quoting capacity into the quiet months deliberately.

Common questions