Who is buying, and what they are actually buying
Mechanical services is among the most actively consolidated trades. The buyer is usually a platform assembling regional density, and it is not buying a truck fleet or a customer list. It is buying recurring service revenue and the crews to deliver it.
That shapes everything about how it reads your business. Replacement work is real revenue, but it arrives when equipment fails and it does not repeat on a schedule. A maintenance agreement base does repeat, and a buyer will pay differently for the two. Which is why the published ranges for this trade look the way they do.
Two publishers, the same measure, bands that barely meet
Unlike some trades, the publishers here agree on the measure. Both report against owner earnings, so the figures are at least comparable.
ClearlyAcquired, published 16 January 2026, states that smaller operators typically see multiples ranging from 1.99x to 3.33x owner earnings.
CT Acquisitions, verified 5 June 2026, publishes 3x to 5x owner earnings for HVAC, with a premium tier of 5x to 5.5x, scoped to owners with $200,000 to $2,000,000 of owner earnings.
Those two bands touch. They overlap between 3x and 3.33x, which is about a third of a turn. Measured against the narrower of the two ranges, they share roughly a quarter of it. Our threshold for treating two sources as agreeing is half the narrower band, so this does not clear it, and the estimator reports the vertical as carrying conflicting evidence rather than merging the bands into one. What it leads with instead is the sold figure on its own, with the higher advisory band set against it as an asking versus sold contrast. That is not a compromise between the two. It is a refusal to make one.
A third source, reporting the first one's numbers
There is a third figure in the record, and reading it carefully is more instructive than adding it to the pile. BizBuySell transaction data covering 2021 through 2025 reports that the middle half of HVAC businesses recorded as sold fell between 1.99x and 3.33x owner earnings.
Those are ClearlyAcquired's numbers. Not close to them, the same, to two decimal places, in both directions.
Two publishers arriving independently at identical figures to the hundredth of a turn is not something that happens. The straightforward reading is that ClearlyAcquired is restating marketplace transaction data. It does not say so, which is the part worth noticing: nothing on the page tells a reader that the figure they are treating as a second opinion is the first one wearing a different name.
We count them as separate sources anyway, because our rule for independence is what a publisher states about its own derivation, and neither states one. We would rather leave the rule alone and tell you it produces a soft count here than adjust it quietly to fit a case. But an owner treating this trade as having three sources should know that it has, in substance, two positions: a transaction record and an advisory band sitting above it.
That pattern is not confined to HVAC. Across every trade in this series where both kinds of figure exist, the advisory number sits above the transaction number, in the same direction each time.
The gap is not noise, and that is the useful part
It would be easy to read two nearly touching ranges as measurement error and split the difference. That reading is available and it is wrong.
Look at what each publisher says it is describing. ClearlyAcquired attaches its lower band to smaller operators, and notes that businesses under $1 million in revenue often trade closer to 2.0x while those above $2.5 million commonly reach 3.0x or higher. CT Acquisitions attaches its higher band to a stated range of owner earnings, and its top of range drivers are the things that make revenue recur.
Read together, the two publishers are not disagreeing about the same business. They are describing businesses at different stages of the same trade. One is a contractor who sells replacements and services what breaks. The other has a maintenance agreement base that produces revenue on a schedule.
So the gap is informative. It tells an owner that where they land is not really a market question. It is a question about their own contract base, which is something the record either proves or does not.
One publisher, three different figures
There is a further complication worth knowing before you quote anything from a single page. ClearlyAcquired publishes more than one number for HVAC in the same place.
Alongside the owner earnings range, it publishes an EBITDA range of 4x to 10.8x for HVAC businesses, and a summary table giving an average earnings multiple of 2.68x to 2.75x with a median sale price of $800,000.
Those are not contradictions. They are different measures and different cuts of the same market, sitting on one page. But an owner who takes the highest number they see and applies it to whichever earnings figure they have to hand will produce something no buyer will recognise. The measure has to match the multiple, and the population has to match the business.
There is a size where the buyer changes, and the measure changes with it
CT Acquisitions makes an observation on its own page that applies well beyond HVAC. Above roughly $1.5 million of normalized earnings in most industries, the buyer pool shifts to lower middle market private equity platforms that price to EBITDA at higher multiples.
That matters here because it means the small business ranges published for this trade describe individual buyers and SBA borrowers. They do not describe the consolidator that may be calling you. Different buyer, different measure, different price.
An owner benchmarking against a published owner earnings multiple while negotiating with a platform is comparing themselves to somebody else's transaction.
Why we do not split the difference
Averaging these two ranges would produce a figure neither publisher reported, and it would erase the one genuinely useful signal in the data, which is that the two bands describe different kinds of business.
Verelume's estimator leads with the sold figure alone and shows all three beneath it, naming each publisher, stating each scope, and carrying each date. Where sources disagree, we report the disagreement. That is the same standard we apply to a business record: when two documents conflict, the answer is to surface the conflict, not to pick the more convenient one.
What a platform buyer examines
Diligence in this trade concentrates almost entirely on whether the recurring revenue is real and whether the crews to service it will stay.
- Active maintenance agreements, counted and read, not estimated
- What each agreement actually commits to, including visits, parts, and response times
- Renewal and lapse history, and how many agreements quietly went dormant
- Recurring service revenue separated cleanly from replacement and project work
- Outstanding warranty and callback obligations on past installations
- Technician certifications, licensing, and tenure
- Commercial customer concentration, and whether those relationships are contractual
- Pricing history, and whether agreement rates have kept pace with cost
What to do before the platform calls
The work that moves a business from the lower band toward the higher one is operational, and it takes time. But the work that protects whatever position you have already earned is a record problem, and it is faster.
A buyer will ask how many maintenance agreements are active. If the answer comes from a spreadsheet that disagrees with the accounting system, which disagrees with the signed agreements in the filing cabinet, the number stops being an asset and becomes a risk. Owners lose value here not because the agreements are weak but because they cannot be proven.
Verelume reads contracts, financials, and operating records and reports what the record supports, where documents conflict, and where a claim has no documentation behind it. Applied to a service agreement base, that means knowing which agreements are current, what each commits to, and whether what you bill matches what you signed, before a buyer is the one checking.