What does per-seat software actually cost you over five years?
Most field-service and trades software bills you per seat, per month, forever. This tool compares that cost against a one-off owned system, using numbers you control. It is not rigged to make custom software look better — read the "what this doesn't capture" section before you trust either number.
Your assumptions
Change any of these. There is no hidden default working against you.
Office staff and field technicians who would each need a licensed seat.
Default of $30/seat approximates a mid-tier field-service SaaS plan divided across a 5-seat crew (checked against Jobber and Housecall Pro public pricing, 2026-08-28 — see sources below). Some tools price per-seat directly, others per-plan with a seat cap; enter whichever your quote actually is.
Most per-seat pricing scales linearly with headcount. Set to 0 if you expect to stay the same size.
Capped at 15 years. Software this old has usually been rebuilt anyway.
We don't publish a fixed price list — see how pricing works for what actually drives this number. $15,000 is a placeholder for a scoped, single-workflow build; a multi-role system with a legacy migration or an accounting integration costs more. Replace it with a real quote once you have one.
A custom build is not free to run forever. $2,000/year is a placeholder for occasional fixes, small refinements, and hosting; a system sharing already-owned infrastructure can run near-zero in hosting alone, but someone still has to maintain the code. Raise this if you expect ongoing feature work, not just upkeep.
Over 5 years
Cumulative per-seat subscription
$12,600
9 seats by year 5
Cumulative owned system
$25,000
Build cost plus 5 years of maintenance
No break-even inside 5 years. At these numbers, the per-seat subscription stays cheaper for the whole period you compared. Try more years, more seats, or a higher per-seat price to see when — if ever — that changes.
| Year | Seats | Subscription (cumulative) | Owned system (cumulative) |
|---|---|---|---|
| 1 | 5 | $1,800 | $17,000 |
| 2 | 6 | $3,960 | $19,000 |
| 3 | 7 | $6,480 | $21,000 |
| 4 | 8 | $9,360 | $23,000 |
| 5 | 9 | $12,600 | $25,000 |
Where the default numbers come from
The calculator does not need JavaScript to load with sensible starting values, but you should replace every one of them with your own numbers before trusting the result. Here is what the defaults are based on, so you can judge whether they're a fair starting point for your shop.
- Per-seat price. Public field-service SaaS pricing pages, checked 2026-08-28: Jobber's Connect plan runs roughly $105-149/month for a 5-user bracket on annual billing, Housecall Pro's Essentials tier roughly $149-189/month for up to 5 users. Divided across a 5-person crew, that lands near $20-37 per seat per month, which is where the $30 default comes from. Quote-based tools such as FieldPulse and Zuper were reported in the $300-600/month range at 3-5 seats in third-party estimates, not vendor-confirmed pricing pages, so treat those as directional only. ServiceTitan is a different tier entirely: roughly $1,000-2,000+/month before add-ons, plus an additional implementation fee and a multi-year contract that reviewers commonly describe as substantial, though we couldn't independently verify a specific fee amount or contract length. That combination is disproportionate for a shop under about 10 seats and shouldn't be used as this calculator's default.
- Build cost. ArborAI does not publish a fixed price list, on purpose — see how pricing works for why. The $15,000 default is a placeholder for a single-workflow build, not a quote. A system with multiple staff roles, a legacy-data migration, or an accounting integration (QuickBooks Desktop over the Web Connector, for example) costs more than a standalone scheduling tool. Two real engagements behind this site ranged from a scheduling-and-dispatch build with a Desktop QuickBooks integration, to a full quoting-invoicing-and-document-archive system that also migrated roughly 42,000 legacy files and reconciled a roughly-490-row accounts receivable spreadsheet before it could go live. Those are not the same price, and neither should be read off this calculator as "the" number.
- Annual maintenance. The $2,000/year default is a placeholder, not a measured figure. One real build was designed to run on infrastructure the client already owned, sharing a server used for other work under an enforced CPU/memory cap, which kept marginal hosting cost near zero — but that is a hosting number, not a maintenance number. Someone still has to fix bugs, keep dependencies current, and make small changes as the business changes. If you expect ongoing feature work rather than just upkeep, raise this number.
None of these figures are outcome claims. Neither of the two engagements behind this calculator has measured time saved, error reduction, or revenue impact from switching off per-seat software — that data doesn't exist yet, and we're not going to invent it here. What's evidenced is what was built and what it replaced; the cost comparison above is the only evidenced economics claim.
What this calculator does not capture
A calculator that only ever tells you to build custom software is not a useful tool, it's a sales pitch wearing a spreadsheet. Here is what the two numbers above leave out, and why a lower cumulative cost is not the same thing as the right decision.
Build risk
The build-cost input assumes the project ships on scope, on budget, and actually solves the problem. Custom software projects run over. A subscription's price is known before you sign up; a build's true cost is often only known after it's built.
Maintenance is ongoing, not optional
Dependencies age. Browsers change. A new tax rule or a new phone OS can break something a subscription vendor would have already patched for its entire customer base. The annual-maintenance input is a placeholder for this, but it can only ever be an estimate, and it tends to be underestimated by whoever is doing the estimating.
No vendor SLA
A SaaS vendor has an uptime commitment, a support line, and an incentive to keep thousands of other customers happy in the same way you need to be. An owned system's uptime is whatever the team that built it, and the infrastructure it runs on, can actually deliver. That can be made rigorous — real backup and restore drills, resource caps so one workload can't starve another — but it has to be built and proven, not assumed.
Key-person risk
A subscription product is maintained by a company, with more than one engineer who understands it. A custom build is usually maintained by whoever built it, or whoever you hire next to learn a codebase they didn't write. If that person disappears, so does your fastest path to a fix.
A good SaaS product ships improvements you would otherwise pay for
Every month a subscription vendor spends engineering effort on a feature you didn't ask for but end up using is a month you didn't have to fund it yourself. An owned system only gets the features someone specifically pays to build. Over five years, a well-run SaaS product's roadmap can be worth more than its sticker price implies.
The honest version of this page's argument is narrow, not absolute: a subscription is often the better buy for a workflow thousands of other businesses already share. Custom makes sense when you have a handful of specific rules, or a specific piece of existing infrastructure, that no off-the-shelf product handles the way your business actually needs it handled. If you're not sure which one you are, that's a better first question for a call than a quote.