Project tools: the annual discount is not a category norm, it is a per-tier decision
In project and task software the annual discount is treated as a fixed convention and is nothing of the kind: across the tools we can price it runs from about seventeen per cent to nearly thirty-seven, it differs between tiers on the same ladder, and at least one vendor's headline summary of its own discount is wrong in the buyer's favour. Add pages that open already discounted and a vendor that publishes no monthly price at all, and the per-seat figure you compare depends on which state you happened to see.
Paying annually saves you something in almost every software category, and buyers have absorbed a rough figure for what that something is — a couple of months, call it a sixth. In project tools that intuition is wrong by a factor of two at one end of the range. Notion's annual discount works out at just over seventeen per cent on both its paid tiers. ClickUp's is thirty per cent on one tier and nearly thirty-seven on the next. Same category, same billing convention, and one vendor is giving away more than twice as much as the other for the same commitment.
The ClickUp case is worth dwelling on because its page gets its own arithmetic wrong in an unusual direction. The header offers 'save up to thirty per cent with yearly', which is exactly right on its Unlimited tier and understates its Business tier, where the yearly and monthly figures work out at nearly thirty-seven per cent apart. An 'up to' claim is normally a ceiling that the small print walks back; here it is a floor the vendor undersold. We record what the page prints rather than what it summarises, which is how the discrepancy surfaced at all.
That page also does something rarer than it sounds: it publishes both billing figures in its own markup at the same time, so each plan card names a yearly price and a monthly price without anything having to be clicked. Most pricing tables hold one state and swap it when you touch the switch, which means a reader — or a scraper — sees whichever state the page chose to open in and may never learn there was another.
Notion's page is the ordinary case and shows why that matters. It opens on yearly billing, so the per-seat rates a visitor sees before touching anything are already the discounted ones, with the monthly rates behind the switch. Nothing is concealed and the switch is plainly there, but the first figure you read is not the price of a month, and if you are comparing three tabs quickly it is very easy to put one vendor's discounted rate against another's undiscounted one without noticing.
For the tools we can price in every market we publish, the entry figures are these: Monday starts at €12 a seat and reaches €24 on its professional tier, Asana runs from €13.49 to €30.49, Evernote from €12.99 to €21.99, and Notion's business tier is €19.5. Each is a per-seat monthly figure, so multiply by headcount before comparing anything, and check which billing state you are reading.
Linear takes the convention to its conclusion by publishing only the annual state at all. Both its paid tiers are quoted per user per month billed yearly, with no monthly toggle we could find, so there is no month-to-month price to record. We do not compute one by dividing, because the annual state is the discounted one and dividing it would produce a monthly figure cheaper than any the vendor offers.
Our coverage of this category is uneven and it is worth saying exactly how. Of the eleven tools we file under project software, three can be quoted as figures in every edition we publish. Two — Trello and Basecamp, among the better-known names in the category — hold no captured pricing at all. Todoist publishes per-seat rates in a form our pages cannot turn into a figure, and Linear publishes no monthly price for us to record. The rest were read in a single market, so quoting them would put one region's currency in front of another region's readers. That is a thin base for a price comparison, and a reader is better served knowing it than by a table that quietly omits eight of eleven.
The practical rule is short. Work out the annual discount yourself, per tier, by dividing the yearly figure by the monthly one, because the vendor's summary of it may be a rounded ceiling, an understatement, or absent. Then check which state the page opened in before you write a number down. In a category where the same commitment buys anywhere between a sixth and better than a third off, the discount is worth as much attention as the headline rate it applies to.
The discount ranges from about a sixth to better than a third
Just over seventeen per cent on both of one vendor's tiers, thirty and nearly thirty-seven on another's. There is no category convention to rely on, and the spread is wide enough to reorder a shortlist that was built on monthly prices.
It varies between tiers on the same ladder
One vendor discounts exactly thirty per cent on one tier and nearly thirty-seven on the next. Working out the saving on the tier you are actually buying is not the same as reading the saving advertised at the top of the page.
An 'up to' claim can be an understatement
One page offers 'save up to thirty per cent' and delivers nearly thirty-seven on its second tier. Ceilings in pricing copy are normally optimistic; this one is conservative, which is only visible if you compute the discount from the two printed figures rather than reading the banner.
Publishing both billing states at once is rare and useful
One vendor's plan cards carry a yearly price and a monthly price in the markup simultaneously, so the comparison needs no clicking and no trust. Most pages hold one state and swap it, which means whichever state the page opens in is the one most readers will record.
Pages that open on yearly show you a discounted rate first
A page defaulting to annual billing presents its lowest per-seat figures before anything is touched. That is not concealment — the switch is there — but it makes cross-tab comparisons unreliable unless you check the state on each one.
We can quote three of the eleven we file here
Two well-known tools hold no captured pricing at all, one publishes per-seat rates in a form our pages cannot resolve, one publishes no monthly price, and several were read in a single market. They are described or omitted rather than estimated, so our coverage of this category is thin and says which parts.
Questions people actually search
How do I work out the real annual discount?
Divide the yearly per-seat figure by the monthly one and subtract from one, tier by tier. It takes a few seconds and it is more reliable than the banner, which may round, may quote a ceiling reached on only one tier, or may understate. In this category doing it yourself changes the answer by up to twenty points depending on which vendor and tier you are looking at.
Why does the price I see differ from the one you publish?
Most often because the page opened in a different billing state from the one our figure describes, or because you are looking at a per-seat rate and we quote per seat too but you have more than one seat. Every figure we publish names its state in the record; where a vendor publishes only one state, that is all we hold.
Why is there no monthly price for some tools?
Because some vendors do not publish one. At least one in this category quotes every paid tier per user per month billed yearly with no monthly option visible, so there is no month-to-month figure to record. We do not divide the annual rate to invent one, since the annual state is the discounted state and the result would be cheaper than anything actually on sale.
Why are some well-known tools missing from your comparisons?
Two of them hold no captured pricing in our records at all, and two more publish rates in a form our pages cannot resolve into a figure. The first is a gap in our reading and the second is a limitation of ours rather than the vendors'. Either way we would rather name the gap than fill it with a plausible number.
Do you earn a commission from any tool named here?
No. We hold no affiliate relationship with any of the project tools discussed on this page, and this page carries no affiliate links. Our method is on the method page.
Prices resolve from dated vendor records at build time and appear in each edition's own currency; only tools read in all three of our English markets are quoted as figures, which is why several discussed above are described rather than priced. Records cited were read on 2026-09-03. Discount percentages are computed from the two figures each vendor prints for the same tier and are our arithmetic on their numbers, not claims either vendor makes. Of the eleven tools we file in this category, three can be quoted in every edition; the other eight are unpriced, unreadable, single-market or publish no monthly rate, and are named as such rather than estimated. This page carries no affiliate links. Related: the annual discount you cannot check, is that price per seat? and how to read a price record. The method →
Found a price that no longer matches the vendor’s page? Tell us — corrections are dated and stay on the page.