How to read the per-seat table on this site
Every table here has the same columns: published seat price, the billing basis it applies to, the monthly-billing rate, the minimum seats, and the effective cost per seat over three years. The ordering is that last figure, ascending, and nothing else.
Two things mislead a fast reader: a price that only applies on annual billing, and a minimum seat count larger than your team.
The pages on this site are generated from one dataset by one script, so the columns behave the same everywhere. Learning them once makes the whole site readable.
The columns
Published seat price. The figure the provider prints, copied exactly, never converted between currencies, and labeled with the VAT basis on UK plans where the provider states one.
On. Which billing basis that figure applies to — annual or monthly. This column is the one no comparison table carries and the one that decides whether the price is the price.
Monthly billing. The provider's own monthly rate where it publishes one. If it does not, the cell says so; it does not repeat the annual figure.
Min seats. The number of seats billed regardless of how many you use. Where a plan has one, every computed figure on the page already includes it.
Effective per seat. The computed figure: what the plan costs across 36 months, divided by the seats you actually have. Tables sort on this.
As published. The provider's own wording, quoted. Where a page contradicts itself, both halves stay in.
The formula, in full
Nothing else is added. No taxes, no regulatory fees, no handsets, no porting, no overage — what is not in the price lists each one and why it is absent. The method page has the tie-break rules.
Why the ordering is one number
Because a weighted score is a place to hide a commercial interest.
The comparison sites in this market order by a composite that mixes price with criteria nobody outside the company can audit, and at least one of them discloses that supplier compensation may affect placement within it. A single computed sort key removes the possibility: there is no weight to nudge, and anyone can recompute the order from the published dataset and get the same rows back.
Ties break on the published seat price, then alphabetically. That is the whole rule.
The first place a quick read misleads you
A price that only applies on annual billing. Most providers advertise the annual rate, and where both figures exist the monthly one runs about 1.40 times higher. If you intend to pay monthly, the ranking you want is built on the monthly column, and the size of that gap is larger than most of the differences between providers.
Read the "on" column at the same time as the price, always.
The second place a quick read misleads you
A minimum seat count larger than your team. A plan at $29 a seat with a two-seat minimum costs a sole trader $58 a month, which is $58 per seat used, not $29.
Our tables already do that arithmetic — the "effective per seat" column divides by the seats you have, not the seats you are billed for. That is why a plan can look more expensive here than on the provider's own page, and it is the correct comparison for a small team.
When ten seats is the wrong basis
Often. Ten seats is the default here because it is the size where most of these plans are actually bought, and every ranking page states the size it was computed at.
If your real number is three, or forty, the ordering changes. The calculator takes a seat count and re-ranks every plan in the dataset against it, using the same formula and the same published minimums. It runs in your browser and asks for nothing.
The pages this dataset produces
Four kinds, all generated from the same rows:
- Provider pages — every published plan for one provider, with the cost per seat at 1, 3, 5, 10, 25 and 50 seats.
- Head-to-head pages — two providers' cheapest plans across every one of those sizes, plus the point at which the answer changes.
- Alternatives pages — the nearest providers in the same category and market, in cost order.
- Size pages: every plan a business of a given size could buy, in order.
None of them is written by hand and none can be edited by hand: a hash check fails the build if a generated page differs from what the script would produce. If a figure is wrong, the fix goes into the data and every page carrying it changes at the next build.
Who this is not for
If you want to be told which provider to buy, these tables will not do it. They order plans by one published-price computation and stop. They do not know which provider has the better support desk or the fewer outages, and no column here should be read as a recommendation.
If you are comparing across currencies, they will not help either: US and UK plans never appear in the same ranking, on purpose. Why that is is a page of its own.
Questions people ask
Why 36 months? Because a phone system is bought once and lived with, and a three-year window makes the annual-versus-monthly difference legible without pretending anyone can forecast further.
Can I get the underlying data? Yes, at data.json, under CC BY 4.0, with the source URL and check date on every figure. Recomputing the tables from it is the intended use.
What happens when a price changes? The row is re-read and re-dated and every page carrying it is regenerated. Any figure older than 90 days is automatically labeled as not re-checked, on every page it appears on, so a stale number announces itself.
Sources
- All figures and counts on this page come from data.json, which carries a source URL and check date for every row.
- The sort key, the tie-break rules and the full formula are published on the method page.
- Individual provider pricing pages are linked from every row in the provider list.
Written by Kaz (pen name). Figures are copied from providers' published pricing pages on the dates shown. We have not used these products. Corrections: contact form.