Remote access licensing for growing teams: how the maths works

The four numbers

Before you open any pricing page, write down four figures. Console users: everyone who needs to log in, including part-time, out-of-hours and cover staff. Peak concurrency: the maximum number of sessions running simultaneously in your busiest hour, which you can measure from your current tool's logs rather than guess. Managed devices: everything needing a permanent agent, counted properly, including machines nobody sits at. Capability list: unattended access, attended sessions, file transfer, session recording, reporting, API, and any integration you would truly use.

Those four numbers turn incomparable tiers into comparable annual totals. Without them, every quote sounds reasonable.

Where growing teams get caught

Per-seat licensing is comfortable at three technicians and uncomfortable at ten, because cost rises in a straight line with a headcount that also rises. The awkwardness usually arrives before the budget review: an apprentice starts, someone covers out of hours, a second site gets its own part-time support person, and each addition is another full seat for someone who connects for twenty minutes a day.

The second trap is device growth you did not authorise. Estates expand quietly: a new till, a signage screen, a spare laptop that becomes a permanent kiosk. If you pay per device, that drift appears on an invoice; if you do not, it still expands your attack surface, so count it either way.

The third is tier drift. A requirement appears (a compliance need for session logs, a report someone in finance wants) that lives one tier up, and the increment applies to every seat you hold.

Modelling two years, not one

Build the twelve-month total twice for each candidate: at your current numbers, and at the numbers you honestly expect in two years. Vendors are compared on the first and lived with under the second. A model that is cheapest today and second-cheapest at double the headcount may still be right, but you should choose that knowingly rather than discover it at renewal.

Add a sensitivity check. If concurrency rose by one, what happens? If twenty devices were added? If you needed session recording? A quote whose total leaps under a small, plausible change is a quote with a hidden cliff in it.

The annual licence review

Most teams review pricing only when a renewal lands, which is the worst moment: there is time pressure and the incumbent knows it. Put a thirty-minute review in the calendar two months before renewal instead, and bring four things to it.

Actual usage from the last twelve months: peak concurrency, sessions per operator, devices genuinely reached. The operator list, checked for leavers and dormant accounts. The device list, checked for machines that no longer exist. And one current competing quote, built from the same four numbers, which is the only piece of leverage that reliably works in a renewal conversation.

Free tiers, and the line they draw

Every serious vendor in this category offers a free or personal tier, and every one of them draws roughly the same line: personal, non-commercial use, with limits enforced by usage detection, session duration caps, or the absence of unattended estate management and audit logging.

Using a personal tier for business support is a poor trade even where nobody notices. Sessions get cut mid-repair, commercial-use detection blocks a connection at the wrong moment, there is no audit trail if anyone asks who connected to what, and there is no support to escalate to when a critical session will not open. If the work is commercial, price a commercial licence and use the free tier only to evaluate the product.