Peak season math: your call volume triples, your capacity does not
HVAC call volume can run 340% above baseline in peak months, and roughly 73% of annual revenue lands in just six of them. Here is how to plan capacity for the spike.
Every HVAC owner knows summer is busy. Very few have looked at how lopsided the year actually is.
Industry figures put peak-season call volume as much as 340% above spring baseline, with roughly 73% of annual revenue landing in six months — June through August, and December through February. Your business is not a twelve-month business. It is two sprints and two lulls, and your phone system was almost certainly sized for the lulls.
That mismatch is the most expensive structural problem in a small contracting shop, and it is nearly invisible because it only shows up in the weeks when you are too busy to look at it.
Why capacity fails non-linearly
Here is the part that catches people out. If your call volume triples, your missed call volume does not triple. It grows faster.
A single CSR handling a steady eight calls an hour has slack. Every call gets answered because there are gaps between them. Push that to twenty-four calls an hour and the gaps disappear entirely — and every call that arrives while she is on another line is simply gone. You did not lose 3x the calls. You lost every call above her ceiling, which might be 60% of them.
Put roughly:
Capacity is a wall, not a slope. Below it you miss almost nothing. Above it you miss almost everything over the line.
This is why contractors describe July as “chaos” rather than “busy”. It is not that the work increased. It is that the failure mode changed.
The four things that break, in order
When we look at contractor call logs across a peak week, the same four failures show up in the same order.
1. Both lines busy during the day. The largest single bucket in peak season, and the one almost nobody plans for. Everybody buys after-hours coverage; the calls are being lost at 2pm.
2. Hold abandonment. Technically answered. Abandoned at 40 seconds. Counts as a miss in every way that matters to revenue, and does not appear in most “answered calls” reports.
3. The 5pm to 8pm gap. Peak emergency density colliding with the moment your office closes. Covered in more detail in the after-hours playbook.
4. Saturday. Weekend replacement quotes and emergency repairs, against the lowest staffing of the week.
Only one of those four is solved by hiring a night answering service, which is what most shops buy first.
What the spike is actually worth
Run this on your own numbers rather than trusting anyone’s benchmark.
Take a three-truck shop with a normal week of 40 calls that goes to 120 in a heat advisory week. If the CSR ceiling is around 70 answered calls a week, that leaves 50 unanswered calls in seven days. Peak-season calls skew toward emergencies, so both the close rate and ticket are higher than your annual average — call it 40% conversion at a $520 emergency ticket.
50 × 40% × $520 = $10,400 — in one week.
Even if that model is generous by half, a single heatwave week is costing more than a year of any coverage solution on the market. And there are typically six to ten of those weeks in a year.
That is the real argument for capacity. Not the annual average — the weeks where the wall gets hit.
Planning for it without hiring for it
Staffing for peak means carrying that cost through March, when you do not need it. Most shops cannot, which is why they under-staff and eat the losses instead.
The alternatives, roughly in order of cost-effectiveness:
Elastic overflow, not headcount. Whatever you use for overflow needs to answer the eleventh simultaneous call as readily as the first. Human services have a headcount ceiling too — during a regional heat advisory, your calls queue behind every other client’s calls. Ask any vendor what their July hold time looks like, not their March average.
Avoid per-call pricing. A metered service bills you hardest in exactly the weeks you are already stretched. See the answering service cost breakdown for how the tiers price the spike.
Triage before you dispatch. In peak weeks the constraint moves from answering to scheduling. A caller who is told honestly that the first slot is Thursday will often take Thursday. A caller who is told nothing calls someone else.
Buy before the season, not during it. Some services need two to four weeks of onboarding. Buying coverage in the second week of July means help arrives in August.
The winter version is the same problem
Everything above applies to the first hard freeze, with two differences: the calls are more urgent because no heat is a safety issue faster than no cooling, and the season arrives on a specific unpredictable night rather than building over weeks.
Contractors who prepare for summer and improvise in December leave the same money on the table twice a year.
The one thing to do before next peak
Pull your call detail record for your busiest week last year and your quietest week. Count calls under 15 seconds in each.
The difference between those two numbers is your capacity wall, expressed in dollars. It is usually the largest single number in the business that nobody has ever written down. Here is how to run that calculation properly.
Then decide what to do about it in April, not in the middle of a heat advisory.
If you want to hear what elastic coverage sounds like, our demo line is (260) 264-9632. It answers as a fictional HVAC company, and it will answer the same way whether it is the first call of the day or the fortieth at once.
Related reading: The five-minute rule · Six phone metrics worth tracking · HVAC answering service cost in 2026
Stop losing the calls behind this problem.
Forward your line tonight and see what a 24/7 answer rate does to next week’s schedule. Fourteen days free, no contract.
