← NeuralBuilt blog
Ownership economics

The Seat Tax

Trey Mossman · August 2026 · every number sourced below

Every tool you rent charges you per seat. Hire someone and your software bill goes up. Same product. Same servers. Nothing changed except your team grew, and the vendor gets paid for it.

I started calling this the seat tax. Once you see it you can't unsee it.

Here's what it looks like for a real service company. I priced the top tiers on purpose because when we build for a client there are no tiers. Everything's included. So the honest comparison is against the best the vendor sells.

A shop with 20 techs runs ServiceTitan at roughly $245 to $500 per tech per month. Implementation alone runs five figures. That's $59k to $120k a year before anyone answers a phone. Add a receptionist at $60k fully loaded. Add scheduling and transcription seats at $40 a user. You're past $200k in year one, and every hire you make from here raises the number.

Team sizeRented stack, per yearOwned build
1 to 5 people$13k to $18k and grows per hireflat
5 to 10$25k to $40kflat
10 to 30$75k to $130k+flat

Now the other line. Sub $5k in hardware runs the whole thing. I know because my own company runs on a Mac mini sitting on a desk, a full CRM database plus a dozen other services on 12GB of a 926GB drive. Ten thousand customers is small data. A one time build plus a retainer for monitoring and upkeep. The line stays flat when you hire. That's the whole argument. One line moves with headcount. One doesn't.

Where this honestly breaks: industry data puts custom build breakeven at 14 to 28 months for most mid size deployments. The under 12 month version only holds when the build replaces a real stack, three or more tools plus a payroll role like inbound calls. That's exactly the kind we do. If someone quotes you a fast breakeven on replacing one $50 subscription, walk away.

I want to be straight about where this breaks, because the research my system pulled says the generic version of this pitch oversells it. And the failure stats are real. Most IT projects miss their objectives, and 60 percent of software cost lives in maintenance, not the build. That's not an argument against owning. It's an argument for vetting whoever builds it exactly as hard as you'd vet the SaaS company. Most people vet neither. They trust the bigger logo.

Build or rent is the easy question. Assets win when the asset meets or exceeds what you were renting, for the reason you rent it. The hard questions are the quality of your builder, the quality of your current vendors, and whether anyone wrote down what the software actually has to do. Get those three right and the seat tax is optional.

Check my math below. If it's wrong somewhere I want to know.

Sources

ServiceTitan pricing (user reported, no published list): Projul, ITQlick, Tooled Up Pro

Calendly pricing: Zeeg, Costbench · Otter.ai pricing: Sonix, tl;dv

Loaded cost of a customer service rep ($60k to $65k with taxes, benefits, training, turnover): BLS data via Salary.com, eesel

Breakeven ranges (14 to 28 months generic; under 12 for multi-tool plus payroll replacement), IT project failure rates, and the 60 percent maintenance figure: NeuralBuilt research review, August 2026, drawing on McKinsey, IEEE, and industry TCO studies.