You close the fourth deal and the growth stops.
Not the revenue. The revenue is bigger, obviously. You just bought it. What stops is the kind of growth that came from buying, because that kind finishes the day the deal closes. From there it is the same trucks, the same techs, the same calls, and a board that would like to hear what organic looks like.
So you go looking, and the first thing you find is that two branches in the same metro are not worth the same per visit. Comparable trucks, comparable crews, comparable call volume, materially different revenue. You ask why. Each manager has an answer, and not one of them survives a follow-up question.
Why doesn't same-store revenue move after the acquisitions?
Because the easy growth got spent on the way in.
An acquisition adds revenue once. The multiple you paid already assumed you would do something with it afterward. And every obvious lever for doing that costs money the thesis has committed elsewhere: more trucks, more marketing spend, more technicians in a market where finding a good one takes most of a year. The Bureau of Labor Statistics projects the trade will need roughly 40,000 new HVAC workers a year through 2034, most of them just to replace people retiring or leaving the field.
Which leaves the calls you are already running. Same truck rolls, worth more each. That is the whole assignment, and it is harder than it sounds, because nobody in the building can tell you where the additional value is currently going.
Why is one branch worth more per truck roll than another?
Put one visit at your strongest branch beside one at your weakest. Same job, same duration, same competent technician.
At the strong branch, the tech noticed something outside the scope of the call, somebody wrote it down, and it became a quote. At the weak branch, the tech noticed the same thing, mentioned it to the homeowner on the way out, and drove to the next call. Nothing else differed.
Why an individual observation dies in a driveway is its own subject, and why field service techs leave revenue on the table walks through it properly. What matters at your scale is a different fact about the same failure: it is not evenly distributed. It clusters by branch.
Why don't standardization programs close the gap?
You have probably run at least one of these. They follow a pattern.
The SOP rollout writes down how a visit should go, distributes it, and collects signatures. The playbook copies the strong branch's process and ships it everywhere else. The scorecard puts a quoting metric in front of every manager each month. The GM transplant moves your best branch manager into your weakest branch and waits.
Some of this helps at the margin. None of it closes the gap, because all of it standardizes process, and the thing that varies is output. You can specify that a technician should document what they find. You cannot verify that they did, because the only evidence would be the record that failed to get created. So the programs land hardest on the branches already doing it, where they change nothing, and slide off the branches that are not.
The GM transplant is the tell. Move the good manager and the numbers tend to follow them, which everyone reads as proof it was a people problem all along. Read it the other way. If the outcome depends on which human happens to be standing there, you do not have a system. You have a person, and a person can only be moved once.
Why can't you see the gap from the P&L?
You can see it. That is what makes it maddening.
The spread is right there. Revenue per visit by branch, ranked, every month. You know exactly which branches are behind and by how much. What the P&L will not tell you is why, and the distance between those two things is the entire problem.
A number that identifies the branch without explaining it routes you straight into anecdote. You call the manager and get a reason. Softer market. Older housing stock. Two techs out in March. A rough quarter. Some of it is true. All of it is unfalsifiable, because there is no record of what the technicians saw and never quoted, so nothing distinguishes a branch with genuinely less opportunity from a branch leaving more of it in the driveway.
More often than not you end up managing the story rather than the branch. And the story is always told by the person whose numbers it explains.
What would it actually take to raise the average branch?
Stop trying to turn average managers into good ones. That is a real project with a real timeline and it is not the short path.
The shorter path is making the outcome stop depending on them. Today, whether a technician's observation becomes a quote runs through whoever is running that branch this quarter. Their habits. Their attention. How hard they push, and whether they were out in August. That is not a standard. It is a personality, repeated by luck, and it explains why your best branch cannot be copied by circulating its playbook.
The change worth making is narrower than a transformation program. Capture has to happen the same way at every branch no matter who is managing it, which means it has to come out of something the technician already does rather than something a manager has to keep enforcing. Get that right and your weak branches stop being weak for reasons nobody can name, and your strong branch stops being a story about one exceptional individual you cannot clone.
What should a multi-location operator require of a platform?
Three things, and they are specific to running more than one shop.
It has to work on whatever each shop already uses. Consolidating acquired companies onto a single field service system is its own multi-year program, with its own budget, its own disruption, and its own decent chance of stalling. Anything that requires that consolidation first is not an organic growth project. It is a dependency on one.
It has to attribute recovered work to a technician and a location. Lift you cannot attribute is lift you cannot defend. When diligence asks where the organic growth came from, "our capture improved" is a sentence. A named technician at a named location attached to a specific closed estimate is evidence, and the difference between those shows up in the room.
It has to be provable at one branch before it becomes a portfolio commitment. You should be able to connect a single shop, run it, and look at real output before anything gets signed platform-wide. A vendor confident in their mechanism will prefer it that way. One who needs the whole portfolio up front is asking you to carry a risk they are not willing to.
Where Mentat fits
Mentat is field service revenue recovery software. It rides on top of the field service system a company already runs, and it is never the system of record: not a field service management system, not a CRM, not a replacement for anything you have. A technician records a short voice memo about what they saw. Mentat structures that memo, prices the work, and drafts the estimate straight back into that same system, where a person on your team reviews it and sends it. Every opportunity carries the name of the technician who spotted it and the location it came from. Today the loop runs against Jobber.
At more than one shop, here is the honest shape of it right now. It runs per shop, on the system that shop already runs, one connection at a time. You can start with a single location, look at what comes back, and expand from there. Managing multiple locations from one review queue is in development and will launch soon.
The architecture underneath is a decision already made rather than a screen you log into. Each shop stays its own tenant rather than a row in someone else's account, and a location becomes real by connecting that shop's system rather than by being typed into a form.
FAQ
How can a PE-backed HVAC platform grow same-store revenue? By getting more out of the calls it already runs rather than buying more calls. The largest recoverable difference between a platform's strongest branch and its average one is how reliably a technician's observation becomes a written, priced estimate, and that varies by branch because it usually depends on the manager rather than on a system.
How do you increase revenue per technician across multiple locations? Not by asking technicians to sell harder. Revenue per tech rises when the work they already notice stops evaporating between the driveway and the office, and it will keep rising unevenly across a platform until capture happens the same way at every location.
How do you standardize service quality across acquired trades companies? Process standardization, meaning SOPs, playbooks, and scorecards, specifies how a visit should go but cannot verify what a technician observed and never wrote down. The part that transfers across brands and systems is standardizing the output, meaning the record each visit leaves behind.
Does this require consolidating every location onto one FSM? No. Mentat rides on the field service system each location already runs. Jobber is live today; ServiceTitan and Housecall Pro are coming.
Can you manage multiple locations from one place today? Not yet. Today Mentat runs per shop, one connection at a time, and you can start with a single location and expand from there. Managing multiple locations from one review queue is in development and will launch soon.