Is that price per seat? The plan named after you is often the expensive one

Twenty-six plans in our record carry a price qualified per user, per seat or per person, spread across ten vendors and six named categories. On those, the figure on the card is a rate rather than a bill, and multiplying it by your headcount is only the start of the calculation. The part that decides the answer is usually the seat minimum, and that is the number vendors are least likely to print beside the price.

Scribe's ladder is the clearest demonstration, and it is not hidden, merely unremarked. Its Pro Personal tier and its Pro Team tier are both priced per seat, and Personal costs slightly more than twice as much per seat as Team does. That alone would be strange enough. What makes it consequential is the minimum: Team must be bought five seats at a time, Personal one at a time. So Personal is the cheaper choice for one person, still cheaper for two, and from the third person onward buying the full five-seat Team minimum, most of it unused, costs less than three Personal seats. The plan named for the smaller buyer is the expensive one from a headcount of three.

It is worth being precise about why that happens, because it is not a trick. A vendor sets a low per-seat rate for teams and protects it with a minimum so that individuals cannot take it. The minimum is doing the segmenting, not the rate. Once you see that, the question to ask a pricing page stops being 'what does a seat cost' and becomes 'what is the smallest cheque this tier will accept', which is a question most comparison tables never ask.

The same shape turns up in consumer software, where it is friendlier and just as easy to miss. NordPass charges £1.09 a month for one person and £1.89 for a family. LastPass prices its individual tier at £2.6 and its family tier at £3.4. Proton Pass asks £3.99 for one person and £5.59 for a family. In all three cases the family plan costs less than two individual licences, so the crossover is not at four people or five: it is at the second person. Two people who each pay for their own licence are already paying more than one family plan would cost them, and that holds in all three of the markets we price these in.

That conclusion survives a gap in our own record, which is worth saying plainly rather than glossing. None of those three records carries the number of people its family plan covers. The prices were read; the seat allowance was not, because it was not printed where the prices were. So the record can tell you the crossover happens at two people, which follows from the prices alone, and it cannot tell you where the saving stops growing, which would need the cap. A per-seat price without a seat count is a rate without its unit, and we would rather publish the half we read than complete it from memory.

One caution belongs beside that arithmetic, because in this category the advertised annual figure is not always the standing one. At least one major password manager publishes a personal annual rate that its own small print describes as a limited-time discount for new customers in the first year only, with the standing rate struck through beside it in the page's markup; the standing rate is a third higher on both its personal tiers. The crossover conclusion survives that — the family plan still costs less than two individual licences at the standing rates as well as the promotional ones — but a comparison quoting the advertised figure is quoting year one, and the arithmetic worth doing is the one on the price you will pay from year two.

Comparing per-seat prices between vendors is harder than comparing them within one, and two ladders in our record show why. Figma prints its rungs on different billing bases side by side: the Professional tier is displayed as a monthly rate while the tiers above it are displayed billed annually, and neither plan's other state appears anywhere on the page. A reader comparing the first figure with the second is comparing a monthly price with an annual one and has no way to know it from the card. Linear publishes only the annual state, quoting both paid tiers per user per month billed yearly, with no toggle we could find; its monthly price is therefore not recorded, and was not back-computed, because inferring a monthly rate from a discounted annual one would invent a number in the vendor's favour.

So the honest comparison procedure is duller than a table and more reliable. Establish for each vendor whether the figure is per account or per seat. Establish the minimum. Establish which billing state the figure is quoted in, and whether the other state is published at all. Only then multiply. Two of those four steps are usually answerable from the pricing card, and the other two frequently are not, which is the actual reason per-seat comparisons go wrong rather than any deception.

One last asymmetry is worth carrying into any negotiation. Where a vendor prices per seat and enforces a minimum, the effective price of your first seat is the whole minimum, and the marginal price of your next is the rate on the card. Teams that grow into a tier get steadily better value; teams that sit just above a threshold pay for capacity they will not use. If your headcount is close to a minimum from below, the seat rate is not what you are deciding about.