ServiceTitan, or something built around you?

Four things, and none of them are close. Breadth — a decade of features I can't match in a twelve-week V1. Availability — you can be using it today, while a custom build has a scoping and build window in front of it. Support — a staffed desk answering at 2am, where I'm one person with a phone and a written response time. Integrations — the connections you need are already built and already maintained by a team. Add a large user community and training material that already exists. If your operation fits their model, those advantages are worth more than a perfect workflow fit, and you should buy it.

They quote per seat after a demo and don't publish a number, so I'm not going to invent one to beat. Get your quote in writing, multiply it by the seats you'll have in three years, multiply that by sixty months, then add implementation, training and the staff time the rollout eats. Compare that total against a one-time build plus hosting. That arithmetic is the only version of this comparison that means anything, and it's yours to run — not mine to guess at.

$25,000 to $65,000 for most V1 builds, with $25,000 the realistic floor for an operations or field service system. The number is fixed in writing before any code exists, along with the timeline and an explicit list of what isn't in V1. Bigger scopes cost more; I'll tell you which parts to cut to hold a number rather than quietly billing past it.

Probably buy something first — and it might not be ServiceTitan. If your job stages are standard and your crew is small, Jobber or Housecall Pro will get you off spreadsheets for a fraction of the effort and a fraction of the price. Custom earns its cost when a tool actively fights how you work, not when you simply don't have one yet. Going straight from a spreadsheet to a custom build is usually paying to automate a process you haven't finished figuring out.

Usually yes — customers, jobs, invoices and history come out through their export and API, and I map them into the new system before cutover. Be realistic about it: a real migration takes weeks, some historical detail never comes across cleanly, and running both systems in parallel for a period is normal. I'll tell you what will and won't survive the move before you commit to it.

Yes, and it happens. If your operation matches the model, you need staffed 24/7 support, or you need to be live next week, buying beats building and I'd rather lose the invoice than sell you a project you'll resent in a year. What I ask for is thirty minutes to hear how you actually work before either of us decides.