An HVAC marketing report is the monthly accounting of what your marketing produced: the calls and booked jobs it generated, what all of it cost, and what the month's test changed.
It is also the fastest way to judge an HVAC marketing agency, because a report without booked calls in it cannot justify its own fee.
Most reports report activity, clicks, impressions, posts published, and leave out the lines that decide whether the fee earned itself, so this page is the line-by-line list to demand, written for the owner reading it rather than the vendor writing it.
What an HVAC marketing report should show
The short answer to what an HVAC marketing report should show is six lines, each with a definition written down once and never quietly changed.
- Booked calls and booked jobs by source, matched to a phone log you can audit.
- All-in spend by channel: fees, software and tracking numbers, not just the ad bill.
- An ROI line on contribution, not revenue, with the margin used stated.
- This month next to the same month last year, or against the degree days recorded.
- The month's named test, its sample size, and its result, win or lose.
- If Local Services Ads are in the mix, leads Google charged next to valid leads.
The HVAC marketing metrics at the top of the report
HVAC is phone-led, so the top of the report is the phone: calls by source, how many were bookable once spam and misdials come out, and how many booked.
Next to them sit the jobs those calls became: estimates set, estimates presented, replacements sold, because booked installs are the line that pays the fee.
The formulas for those HVAC marketing metrics are in HVAC KPIs, and the report should carry those same written definitions, so a rising number is never a moving one.
If the report includes recorded calls for lead-quality review, consent rules attach: federal law (18 U.S.C. 2511(2)(d)) lets a party to a call record it, state law can be stricter, and whether a tracking vendor fits the federal exception is live litigation, so announce recording on every call.
This is not legal advice, and your state licensing board or attorney confirms the rule where you operate.
Spend, and an ROI line built on contribution
The spend line is everything you pay for marketing, by channel: the fee, the ad spend, tracking numbers, landing pages and software.
The return line has a standard definition: the Marketing Accountability Standards Board's dictionary defines marketing ROI as the contribution to profit attributable to marketing, net of the marketing spending, divided by the marketing invested.
Contribution is the word that does the work: a report line that reads $10 of revenue per $1 spent is return on ad spend, a different ratio that ignores what the sold job costs to deliver.
HVAC marketing ROI runs that formula on real numbers, including the phone attribution that makes it computable.
A report worth its fee also watches repeat work: the same dictionary defines retention rate as the ratio of retained customers to the number at risk, and customer lifetime value as the dollar value of a customer relationship based on the present value of its projected future cash flows.
An HVAC version of that math is illustrative arithmetic on your own books, because no sourced HVAC lifetime value exists.
Compare like months, or the season reads as a result
Month-over-month arrows are the most common lie in HVAC reporting, and the weather is why: in 2025, June through August held about 64% of US cooling degree days and December through February about 59% of heating degree days, according to EIA's Monthly Energy Review.
US cooling degree days by month, 2025
A report that puts October's booking drop under a red arrow is mostly reporting the weather, and the honest comparisons are this July against last July, or this month against the degree days it actually recorded.
Degree days measure weather-driven need rather than service calls, one year is one sample, and a national average hides your region, but that curve is why the report needs a like-for-like rule.
The named test and its result
A report from a conversion-focused provider should name the month's test and its result, and the result should be allowed to be a loss.
Evan Miller's A/B testing primer is why: significance is only valid if the sample size was fixed in advance, and stopping when a result looks significant can push the false-positive rate to 26.1% in his worst-case example.
As our own rough arithmetic, detecting a lift from 5% to 6% conversion takes roughly 8,000 visitors per variant, about 16,000 in total, and HVAC's seasonal traffic swings add confounding on top.
The honest shape of HVAC testing is few, big tests, not a stream of small winners.
Two lines that need their own definitions
If Local Services Ads are part of the spend, separate what Google charged from what you would count as a valid lead: Google says it does not charge for leads it judges invalid or low quality when they arrive, re-checks charged leads with automated models and may credit some later.
Google's list of valid leads includes a message, a voicemail, a call you answer, a missed call without a voicemail that you return and reach or leave a voicemail on, and a booking request, and Google no longer credits leads for "job type not serviced" or "geo not serviced".
The website numbers with a named source are Google's Core Web Vitals: a good experience means a largest contentful paint within 2.5 seconds, an interaction to next paint of 200 milliseconds or less and a cumulative layout shift of 0.1 or less, at the 75th percentile of page loads, per Google's web.dev guidance last updated October 31, 2024.
Those are experience thresholds rather than a ranking or conversion promise, so read them as a health line, not a forecast.
How to read this month's report
Score this month's report against the six lines at the top: if they are all there, you can argue about the numbers, and if they are not, you are reading a brochure.
This is also the list More Booked Installs holds itself to, one controlled test a month plus a results report, and the starting point is a free audit that shows where your website leaks booked calls today.
Frequently asked questions
What should a monthly HVAC marketing report show?
Booked calls and booked jobs by source, matched to a phone log you can audit; all-in spend by channel; an ROI line built on contribution rather than revenue; the month next to the same month last year; and the month's named test with its result.
Are clicks and impressions enough to judge a marketing report?
No. They are diagnostics about traffic, while the fee is justified by calls and booked jobs, because homeowners hire an HVAC company by phone. A report can show both, but the results lines come first.
How should a report compare a summer month to a fall month?
Carefully, or not at all: June through August held about 64% of US cooling degree days in 2025, according to EIA, so a July-to-October drop mostly measures weather. Compare like months, or read the month against the degree days it actually recorded.
Should a report separate Local Services Ads leads Google charged from valid leads?
Yes. Google says it does not charge for leads it judges invalid or low quality when they arrive and re-checks charged leads with automated models, so a charged-lead count and a valid-lead count are two different lines.
Why should a report name the month's test?
So you can check the test was run honestly. Significance is only valid if the sample size was fixed in advance, and Evan Miller's worst-case example puts the false-positive rate at 26.1% when a test is stopped the moment it looks significant.