Why Per-Seat SaaS Pricing Punishes You for Growing

Per-seat pricing is the most normal thing in business software and the least examined. Fifty dollars per user per month reads as small, so nobody runs it forward.

Run it forward and it stops being a price. It becomes a fee attached to every person you hire, charged every month, rising every year, for a product that does not change because you hired them. This post does the arithmetic in full, then makes the honest case for when per-seat is the right deal anyway.

Everything below uses one worked example: a single tool at $50 per user per month, with 5% annual price increases. That escalation rate is conservative. Published SaaS renewal increases commonly run higher.

Why does per-seat pricing get more expensive faster than your business grows?

Because two things scale at once and they multiply rather than add. Your seat count goes up with headcount, and the price of each seat goes up at renewal.

A business that grows headcount 10% a year while absorbing 5% annual price increases is not seeing a 15% cost increase. It is seeing 1.10 times 1.05, or 15.5% compounding, every year, on a base that is also compounding. Ten years of that turns a modest line item into one of the larger fixed costs in the business.

The part worth sitting with: none of that spending buys you more software. The vendor ships the same product to a 20-person company and a 38-person company. You are not paying for capability. You are paying for the fact that more people at your company need to open it.

What does one extra seat actually cost over ten years?

About $7,500, on a $50 per user per month tool with 5% annual increases. Five hires is roughly $37,700, in one tool.

That is one tool. Most operating businesses run four to six per-seat products at once: the CRM or field service platform, the comms tool, the document suite, the e-sign product, the time tracker. Every hire is a seat in each of them. The real number for one new person is that table multiplied across the stack.

What happens when headcount growth and price escalation compound?

They stack, and the back half of the decade costs far more than the front half. Here is a 20-person business adding two people a year, on one tool at $50 per seat with 5% annual increases.

Three baselines make the shape obvious. Frozen at 20 seats with no price increases, the ten-year total is $120,000. Frozen at 20 seats with 5% increases, it is about $151,000. Grow from 20 people to 38 and it is about $225,000. Growth alone added roughly $74,000, and the price per seat in year 10 is 55% higher than the price you signed.

Year 10 costs almost three times what year 1 costs. Your revenue may well have grown too. The point is that the software did not.

What does the same growth cost on software you own?

The same, at 20 users or 38. A build has no seat line, so headcount does not appear in the bill.

Using the numbers we quote publicly: a $30,000 build is $30,000 over ten years, and you own the asset at the end. That figure does not move when you hire.

The two lines move in opposite directions. Rented per-seat software gets more expensive per person over time. Owned software gets cheaper per person as you grow, because the cost was fixed when you paid it.

Two honest caveats. Hosting is not literally frozen forever, though adding 18 internal users to an internal tool does not meaningfully move a hosting bill. And support is a real cost we do not pretend away: our model is no flat monthly retainer, so you either take the code and owe nothing further, or you pay hourly, priced by complexity with severity triage. What you never pay is a fee for having more staff.

At how many seats does a custom build actually break even?

Fewer than most people guess, and the answer depends entirely on your time horizon. Against one tool at $50 per seat with 5% annual increases, a $30,000 build breaks even at:

Read that carefully rather than triumphantly. The 8-seat figure is arithmetically true and practically bad advice for most 8-person companies, because it assumes you will still be running the same workflow in ten years and that you can spare $30,000 today. At 20 seats, which is the $12,000 a year case, break-even arrives at year 2.5 and you are $90,000 to $121,000 ahead by year 10 depending on whether prices escalate. That is a decision an operating business can actually make.

This math only holds if the build genuinely replaces the tool. Commissioning custom software that runs alongside the subscription you keep paying is the most expensive outcome available.

When is per-seat pricing genuinely a fair deal?

When each additional seat delivers real additional value or costs the vendor real additional money. Four cases where it clearly does:

• The tool has a genuine network effect. Communication platforms, shared design workspaces, collaborative documents. The tenth user makes the product measurably better for the other nine. You are buying the network, not nine copies of a login. Charging per participant is a defensible way to price that.
• Seat count is low and likely to stay low. Five people on a $50 tool is $3,000 a year. Ten years is about $37,700, well under a $30,000 build plus a decade of hosting. Rent it, and do not think about it again.
• Cash matters more than ten-year cost. Early stage, uncertain product, uncertain survival. A monthly subscription is survivable and $30,000 is not. Renting buys optionality, and optionality is worth paying for when you genuinely do not know what you will need next year.
• The workforce is seasonal or high churn. Per-seat flexes down when you shrink, provided you are not locked into an annual commit with a seat minimum. A contractor running 30 people in summer and 12 in winter gets real value from that flexibility. Check the contract, because seat floors quietly remove the only advantage.

There is a fifth case that is not about pricing at all: commodity workflows. Accounting, payroll, and email are not where you differentiate. Per seat or not, do not build them.

What does seat pricing cost you that never shows up on the invoice?

It makes you ration access, and rationed access has its own price. When a login costs $600 a year, predictable things happen.

• People share credentials, which destroys your audit trail and is usually a contract violation.
• Staff who should be entering data directly send it to someone who has a seat, so one person becomes a human integration.
• New hires wait days for access because someone has to approve the spend.
• Frontline or read-only staff who need one screen get charged for the full product, or get left out and work from a printout.

None of that appears as a software cost. It appears as slower onboarding, weaker data, and time. On a tool you own, the number of accounts is a configuration setting, not a purchase order.

Where does the crossover realistically land?

Four rules that hold up for most small and mid-size operating businesses:

1. Under 10 seats on a given tool: rent. The build almost never pays back inside a horizon you can forecast.
2. 10 to 20 seats and the workflow is specific to how you operate: run the numbers. Five-year break-even sits near 14 seats, so this is where the answer genuinely depends on your situation.
3. Over 20 seats on a tool that does not fit: the case is already made. Three-year break-even is behind you, and every hire widens the gap.
4. Weight it for growth. If you expect meaningful headcount growth over five years, the crossover arrives sooner than today's seat count suggests. Decide on the seats you will have, not the seats you have.

The honest summary: per-seat pricing is fine when it prices something real, and expensive when it prices your payroll. Most operating businesses cross that line somewhere between 15 and 25 seats without noticing, because the invoice only ever goes up a little at a time.

Run it on your own numbers

Nothing above is your number. It is a worked example with the assumptions stated so you can swap yours in.

The fastest way to get your real figure is the SaaS cost calculator. Put in your seat count, your per-seat price, and your expected growth, and it will show you the ten-year line. If you would rather have someone go through the whole stack with you, the free teardown is 30 minutes and produces a written number, including the tools we think you should keep renting. For a fuller version of this exercise across an entire software stack, see the real five-year cost of running on SaaS.