ONE OVERHEAD RATE. THREE DIFFERENT BUSINESSES.
Service agreements, replacement work and new construction have different overhead loads, different billing cycles and different labour. Apply one blended rate to them and the numbers you price from are wrong in both directions at once.
An HVAC contractor at $3.2M with $640,000 of service, $1.28M of replacement and $1.28M of new construction holds about $512,000 of overhead at the trade's 16 percent rate. The true loads are closer to 26 percent on service, 16 percent on replacement and 11 percent on new construction. Applying the blended 16 percent to all three underprices service by 10 points of overhead and overprices new construction by 5, so the strongest division looks ordinary and a weak one looks acceptable. HVAC subs at $1M to $5M average 24 percent gross and 8 percent net before taxes.
Nothing here is a pricing mistake by your estimator. He was given one overhead rate for a company that has three.
This post covers one of the four numbers and how a divisional business breaks it. That page covers all four and how the trade is managed month to month. Read HVAC Operating System for the complete treatment, worked figures included.
WHAT ARE THE THREE BOOKS?
Service agreements are a recurring revenue business. Contracts renew, the customer pays before or shortly after the work, cash cycles in days, and the cost to serve is dispatch, on call rotation, truck stock and scheduling software. Gross margin is high and every dollar of it is expensive to produce because somebody has to answer the phone at 9pm.
Replacement and retrofit is a consumer sales business. It follows the weather, is frequently financed, and depends on lead generation. The cost to serve is marketing, sales commission, financing administration and warranty reserve.
New construction is a contracting business. It is bid, submitted, scheduled, billed on a schedule of values, retained at 5 to 10 percent and collected in 45 to 90 days. The cost to serve is estimating, submittals, project management and the working capital to float it.
Those are three companies. Most HVAC contractors under $10M operate all three and report them as one.
WHY ONE OVERHEAD RATE BREAKS ALL THREE.
Take a $3.2M contractor: $640,000 of service, $1.28M of replacement, $1.28M of new construction. HVAC at $1M to $5M averages 16 percent overhead, so about $512,000 of it, and the natural thing to do is apply 16 percent to everything.
It does not break down that way. Service consumes the dispatcher, the on call premium, the truck stock and the software licence, and its real load is closer to 26 percent, or $166,400 against $640,000 of revenue. New construction consumes the estimator and the project manager, and against $1.28M its load is closer to 11 percent, or $140,800. Replacement is near the blended figure at about 16 percent, or $204,800.
Price service at a 16 percent overhead load and you are 10 points light on the division with the fastest cash and the highest margin, which means you are leaving money on the table on the only work that pays before you do. Price new construction at 16 and you are 5 points heavy on the division that is already the hardest to win, so you lose bids you could have taken.
WHAT BREAK EVEN VOLUME MEANS NOW.
Break even volume is one of the four numbers, and it is the one that goes furthest wrong here. Break even is overhead divided by gross margin percentage, and both inputs differ by division.
Service at a 45 percent gross margin with $166,400 of overhead breaks even at about $370,000 of service revenue. New construction at a 19 percent gross margin with $140,800 breaks even at about $741,000. Those are two different companies with two different answers, and a blended break even of $512,000 over 24 percent gross, about $2.13M, describes neither of them.
The consequence is that you cannot answer the only question break even exists to answer. If service revenue drops 15 percent next quarter, how much new construction do you have to sell to stay whole? With a blended number you can guess. With three numbers you can calculate it.
MOST SUBS MISS THIS: SERVICE IS FUNDING THE OTHER TWO.
Here is the part worth screenshotting. In almost every HVAC company under $10M that has not split its books, service agreements are subsidising new construction, and the P&L is built so that nobody can see it.
Service collects in days and new construction collects in 45 to 90 with retainage behind it. That means service cash is paying new construction payroll every week. The company feels solvent because the bank balance holds, and the reason it holds is that one division has a working capital cycle short enough to cover another division's float.
That is a real business model and plenty of good contractors operate it deliberately. The problem is operating it accidentally. If service is the balance sheet and nobody has said so, the first thing that happens in a soft quarter is that somebody discounts service agreements to hold volume, and the discount falls on the division that was funding everything else.
THE THREE STEPS TO SPLIT IT.
This does not need new software or a second company. It needs three things in the chart of accounts and one decision per overhead line.
1. A division code on every revenue transaction and every direct cost transaction. Service, replacement, new construction. Three codes, applied at entry, no exceptions.
2. Every overhead line assigned to one division or split by a stated driver. Dispatcher to service. Estimator to new construction. Marketing to replacement. Rent by headcount, insurance by payroll, owner salary by time. Write the driver down so next year's allocation is the same allocation.
3. A divisional gross margin and overhead rate reported monthly, and a break even volume per division reviewed quarterly. Three columns on one page.
The first month of this is uncomfortable, because one division is usually worse than anybody thought. That is the point of doing it.
WHAT THIS IS WORTH ON $3.2M.
HVAC at $1M to $5M averages 8 percent net profit before taxes against an 11.5 percent target. On $3.2M that spread is roughly $112,000 a year.
Divisional pricing does not recover all of it, and anybody promising that is selling something. What it does recover is the two mispriced halves: service priced at its real overhead load, which is 10 points above where the blended rate puts it, and new construction bid without absorbing 5 points of somebody else's dispatch cost.
The second one is the sleeper. A new construction division that stops absorbing service overhead can bid 3 to 4 points tighter without touching its own margin, and on a trade where bid outcomes turn on a point or two, that is the difference between a full schedule and a thin one.
THE SHORT LIST.
FREQUENTLY ASKED.
