Running an 81-Year HVAC Company in America's Hardest Market w/ Lawrence Castillo

Lawrence Castillo runs Brody Pennell, an 81-year-old Los Angeles HVAC company operating in a market with no weather, no urgency, and $200 leads. He cut all 25 lead aggregators and call volume did not move. Inside: why the hardest market produces the best operators, what labor really costs, and the recruiting system behind going from 34 employees to 130.

Figuring out how to grow an HVAC business in a competitive market usually starts with the wrong question. Most operators ask how to buy more leads. Lawrence Castillo went the other direction and cut all 25 of his.

Castillo is the operating partner of a portfolio of Southern California home service companies. The platform is Brody Pennell in Los Angeles — founded in 1945, older than In-N-Out, and by his account the largest heating and air conditioning company in LA proper. He took the seat in 2020, when the business had 34 employees, beat-up trucks, few reviews, and a clunky website.

Five and a half years later it has 5,000 Google reviews and a headcount that peaked north of 300 across the group.

“This is the city that consolidation didn’t want to come to because the weather is mild.”

What follows is how a company grows in a market where nothing about the conditions helps.

The Hardest Market Produces the Sharpest Operators

Los Angeles removes every advantage an HVAC operator normally leans on. There are 20 million people within a two-hour drive, and on a 70-degree afternoon almost none of them are thinking about their system.

Castillo describes the daily problem plainly. Fuel runs $6.75 a gallon. Workers’ compensation and windshield time are expensive. And tens of thousands of small contractors are giving jobs away, which sets the price the homeowner expects to hear before anyone knocks.

The absence of weather changes what a technician can do inside a house. In Phoenix a tech can point at a failing amp draw and the customer feels it that afternoon. In West LA, the customer is not running the system, so there is no urgency to convert.

“We have to be better. My technicians have to give a greater level of customer service, they have to be better at their tune-up.”

That is the useful frame for any operator in a crowded market. A competitive market does not require better luck. It requires that every part of the operation carry weight the weather would otherwise carry for you.

The consolidators learned it expensively. Castillo has watched groups buy five brands in Southern California and quietly collapse them into one or two, which is what happens when a playbook built for a market with real seasons meets a market without one.

They Cut 25 Lead Aggregators and Call Volume Did Not Move

Bought leads are rented demand. Castillo’s marketing vendor had 25 lead aggregators running as the answer to a slow board, and cutting all of them produced no measurable drop in call volume.

The numbers behind that decision are worth understanding. A friend with a $50 million company in Fresno pays roughly $20 for a lead from one aggregator. In Los Angeles, Castillo pays $200 for a lead from the same source.

“We were just handing money out the window for calls that we were just paying a fortune for that weren’t even closing.”

The structural problem is not price, it is exclusivity. Shared marketplace leads get sold to three to eight contractors at once, which turns every one into a speed-and-discount race and builds no brand equity on the way through. The cheap lead is frequently the expensive customer.

What replaced it was the database. An 81-year-old company has served a lot of households, and the outbound team works that list rather than paying to be introduced to strangers.

Memberships are the engine underneath it. Brody Pennell runs a deliberately simple plan, two visits a year at $270, three visits at $320 with plumbing, billed monthly, no tiers, and when Castillo arrived they were neither selling nor renewing them. That was the first thing he flagged as an existential problem, and the data backs the instinct: maintenance plan customers carry dramatically higher lifetime value than service-only customers.

He applies the same discipline geographically. A revenue heat map tells him where the brand already carries weight, and every attempt to push into zip codes outside that footprint has cost money. Concentration beats expansion when the brand has not arrived yet. If you are still buying your way to volume, lead generation built on owned demand is the version that compounds.

Run the Backwards Math

A private equity group operating 35 locations toured the business and told Castillo he would buy the company for the call center alone. That is not a compliment about software. It is a compliment about math.

The operating model runs in reverse. This many people require this many calls, which produce this much revenue, and the daily job is closing whatever gap sits between the board and that number.

Analytics make the gap visible. Castillo can walk into dispatch and know where tomorrow stands within ten seconds, though the team is focused on tonight, on making sure no truck is parked on the side of the road in West LA waiting for a call.

The organizing principle came from Leland Smith, who ran Service Champions when Castillo was a general manager there.

“If it’s not in writing, it doesn’t exist.”

Everything at Brody Pennell is written down, which is what makes accountability possible at all. A process that lives in someone’s head cannot be held to a standard.

AI now handles the parts of that system that used to consume payroll. Call dispositioning was once done by a person listening to recordings and flagging opportunities, and booking percentages were unreliable because agents were excusing their own calls. It is automated now, and so is CSR grading, which means the booking number is finally a true number.

Castillo is candid that the customer experience took time to get there. Two years ago customers were openly unhappy with AI answering the phone. Today he says he does not hear complaints, and expects inside headcount to keep shifting as a result.

Labor Is the Variable That Decides Whether You Are Profitable

Equipment and materials can be held to a percentage of revenue. Marketing can be held to a percentage of revenue. Labor is the line that moves, and it is the line that determines whether the year works.

Castillo is direct about the mechanism. Unmanaged overtime erodes margin. A two-day project that becomes a three or four-day project was not sold at that price, and the difference comes out of profit rather than out of the customer’s invoice.

The benchmark data agrees on where the leak sits. Industry guidance puts healthy blended gross margins around 50%, with technician utilization as the primary driver of whether a company hits its net target meaning idle time and job overruns cost more than any line item an owner is likely to be watching.

His broader point is about the pace of change. Equipment costs moved, fuel costs moved, labor costs moved, and the operators still standing were the ones who repriced and restructured instead of absorbing it quietly. Watching the bank balance is not the same as watching the P&L.

Hire a Recruiting Director Before You Think You Can Afford One

The first hire Castillo made after taking over was a recruiting director working 40 hours a week on nothing else. Not a technician. Not a manager. A person whose entire job was filling the bench.

The reasoning was about his own calendar. An operator turning around a business is on the phone with technicians, sitting in the call center, and reviewing job costing in accounting, and recruiting is the task that gets postponed every single day under that load.

With someone owning it full-time, headcount moved from 34 to 50 to 60 to 80 to 100 to 130.

The next step was building their own school. Brody Pennell recruits directly off the street, receives around 5,000 applications per cycle, narrows to 15 or 20 candidates, and runs a three-month program combining classroom work, rooftop systems on their own building, and ride-alongs in actual homes.

The economics are inverted on purpose. A Southern California trade school costs a student $17,000 to $22,000 over ten months. Brody Pennell pays them instead, and gets a technician trained entirely in their processes rather than someone else’s.

Castillo is realistic about attrition. Start with 15, graduate around 13, and hope to still have ten a year later, more likely seven or eight. You over-hire because you are going to lose people, and that is the plan rather than the disappointment.

What made it possible was not a revenue threshold. It was recruiting the person who ran training at the largest local trade school, which took Castillo more than a year of courting to land.

What to Do With This

Three moves worth making this month.

Pull your cost per booked job, not your cost per lead, for every channel you run. The gap between those two numbers is where aggregator spend hides.

Count how many households are in your CRM and what percentage carry an active membership. That ratio is the difference between owning demand and renting it.

Assign recruiting to one person with dedicated hours, even if the business is small enough that the person is you or a family member. Recruiting that shares a calendar with operations never happens.

Watch the full conversation with Lawrence Castillo on TradeOps Radio, and subscribe on YouTube or Spotify for future episodes. If you want to look at what your marketing spend is actually producing in booked jobs rather than leads, book a free strategy call and we will go through the numbers with you.