Per-user vs. per-admin pricing: the math for MSPs

Here is the answer before the arithmetic: an MSP monitoring several client businesses has far more monitored employees than console users, usually by two orders of magnitude. Any model billed on the larger number will cost more. The exception is a single-company buyer with a large management team, where the numbers invert.

The ratio is the whole argument

Three realistic portfolios, using an average of 45 monitored employees per client and the number of technicians who genuinely need a dashboard open:

Note what happens down the middle column versus the right one. Monitored employees quadruple. Dashboard logins double. Console access grows sub-linearly with your portfolio because a technician who can support four clients can usually support six.

The part that is easy to miss

The comparison above is a snapshot. The more expensive difference is what happens over the following year without you doing anything. Under per-user billing, every client's hiring raises your cost. You did not deploy anything new, you cannot invoice more without a conversation, and the increase lands on margin. Under per-login billing, a client growing from thirty to sixty people changes nothing.

This also changes which clients you offer monitoring to at all. When attaching a twelve-person client carries a per-head cost, the deployment effort often exceeds the return and the service quietly does not get offered. When adding a client is free, the calculation is only about whether the service is worth doing.

What actually counts as billable

Read the definition, not the headline number. For SnitchOS the billable unit is an enabled Admin or Manager login, counted on the monthly billing date, whether or not it was used that month. Monitored employees, devices, and clients are included. That means the number you are billed for is one you can look at and count, and deactivating an unused login removes it from the next invoice.

For per-user tools, check whether the count is monitored users, licensed users, or devices, and whether it is a peak or an average across the month. Those three definitions can produce materially different invoices from the same deployment.

When per-user pricing is genuinely cheaper

If you are a single business with 25 employees and eight managers who all need to run reports, per-user pricing will almost certainly cost you less. The ratio is close to 3 : 1 rather than 90 : 1, and per-login billing has no advantage to offer.

The same is true for a very small MSP with one client and several technicians who each want their own console. The model only pays off once you are monitoring meaningfully more people than you have people looking.

The honest limitation

A pricing model is not a product. Cheaper input cost does not help if the tool cannot produce the report your client expects, deploy through your RMM, or scope a manager to one client. Price the models against each other only after both tools have cleared the functional bar you actually need — and check that bar during a pilot, on a real client, rather than against a feature grid.