Every trades owner we talk to has some version of the same worry. Bring in AI, and the crew shrinks. Fewer dispatchers, fewer CSRs, maybe fewer techs once the schedule "optimizes" itself. It's not a paranoid fear. It's the fear the marketing around AI has been selling for two years: do more with less, cut the headcount, protect the margin.
We're not going to tell you that fear is irrational. We're going to tell you the data doesn't back it up.
What the data actually shows
A study using firm-level spend and workforce data across 21,000 US businesses — co-authored by Ramp and Revelio Labs, and cited by Ara Kharazian (@arakharazian) on X on June 30 — found that companies adopting AI heavily grew headcount by 10% over the following two years. Companies that adopted AI lightly, or not much at all, showed no significant headcount change either way.
A few things worth being precise about, because the finding is easy to overstate.
First, this is a broad, cross-industry dataset. It is not a trades-specific or home-services-specific study. Nobody has run this exact analysis on HVAC, plumbing, roofing, or garage door companies specifically, as far as we've seen. We're citing it because it's the best available evidence on the actual relationship between AI adoption and headcount, not because it's about your industry by name.
Second, the effect belongs to heavy adopters, not everyone who adopts AI. Bolting a chatbot onto your website or running a few AI-written ads doesn't show up in this data as growing headcount, because that's not what "heavy adoption" means in the study. Light and moderate adopters were flat. The growth effect is specific to firms that went all-in on AI as part of how the business runs, not firms that treated it as a marketing add-on.
Third, this is a correlation from a large dataset, not a controlled experiment on your specific business. It's a strong data point. It is not a guarantee.
With those caveats stated plainly: the headline finding still directly contradicts the "AI adoption means layoffs" story that's dominating a lot of trades-industry chatter right now. Firms that lean into AI hardest aren't the ones shrinking. They're the ones hiring.
Why this makes sense once you think about what AI actually does well
Here's the mechanism, and it's not mysterious once you separate two very different uses of "AI" that get lumped together constantly.
Use case one: AI as a cost-cutting bolt-on. You keep your existing operation exactly as it is, and you sprinkle AI on top to shave headcount — an AI answering service instead of a night-shift CSR, an AI script generator instead of a copywriter. This is a real use case. It's also a narrow one, and it's the one most people picture when they hear "AI is coming for jobs."
Use case two: AI as an operating layer. This is different in kind, not just degree. When AI tooling is actually embedded in how a business runs — dispatch, follow-up, scheduling, quoting, the stuff that used to bottleneck growth because a human could only juggle so many jobs in flight at once — the constraint that gets removed isn't labor cost. It's throughput. A business that used to turn away calls during a rush, or lose leads to slow follow-up, or double-book techs because scheduling was manual, can suddenly handle more volume without the wheels coming off.
More volume means more jobs. More jobs means more trucks, more techs, more dispatchers to run it. That's the mechanism behind the Ramp/Revelio Labs finding, and it's a plausible one: remove the bottleneck, and growth shows up as headcount, not as the same crew doing the same work faster for the same money.
This tracks with what we're seeing in the marketing-agency corner of this industry too. Firms like Systemized Media and Service Hero, both serving hundreds of HVAC and home-service clients, frame AI and smart tech explicitly as an efficiency layer — shorter follow-up windows, fewer callbacks, better scheduling — not a replacement for skilled labor. That's a narrower slice of the business than a full AI operating layer, but the direction of the argument is the same: AI done well expands what the business can handle, it doesn't just compress the team doing the handling.
The "bolt-on to cut heads" approach and the "operating layer to remove bottlenecks" approach are genuinely different plays with different outcomes. Conflating them is where the layoff fear comes from.
What this means when you're thinking about exit
If you're building toward a sale, here's the connection, stated carefully.
A buyer diligencing a trades business is not looking for a company cut to the bone. They're looking for a business that runs without the owner in every truck and every call, with a team and a culture that will still be functioning the Monday after close. A stripped-down operation with AI replacing half the org chart doesn't read as efficient to a buyer. It reads as fragile, and it raises the question of whether the "efficiency" survives a change in ownership at all.
A business that's grown headcount alongside AI adoption tells a healthier story: the systems are absorbing more volume, the team is scaling with the business, and the operation isn't dependent on one person's hustle to hold together. That's a better signal to walk a buyer through than a shrinking payroll next to a growing AI vendor bill.
To be clear about what this isn't: it isn't a formula, and AI adoption alone doesn't move your multiple. Buyers are still going to screen hardest for customer concentration, fleet age, and how dependent the business is on the owner personally. Those are bigger levers than whether you've adopted AI. What growing headcount alongside AI adoption does is remove a red flag, not add a guaranteed premium. It's one input among several a buyer is going to weigh, not a shortcut past the others.
Where Advisy fits
We build AI as an operating layer, not a bolt-on — the kind of adoption the data above associates with growth, not the kind associated with a quieter phone line and a smaller team. If you're weighing AI adoption against what it does to your team, and eventually to what your business is worth to someone else, that's the conversation we have.
Apply at advisy.com/#apply.