Mirror

The Annual Discount Is a Loan You Make to Your Vendor

Twenty percent off for paying yearly sounds like free money. Priced as what it actually is — an interest-free loan with a cancellation penalty — it is a much narrower deal.

Every SaaS pricing page offers the same trade: pay for twelve months up front and get roughly two of them free. It is presented as a discount, the toggle is usually defaulted to it, and the monthly price is often displayed in grey next to the annual one in black, which is why I rebuild the number myself before reading it. I took the deal reflexively for years. Then I started writing down what I was actually agreeing to, and the shape of it is not a discount.

a hand passing twelve stacked coins across a desk in exchange for a single receipt marked 20% OFF

What you are doing is lending the vendor eleven months of money at zero interest, in exchange for a fee reduction, with a term that says you cannot call the loan back. That framing changes the analysis in three specific ways, and each one has cost me money at least once.

The first is opportunity cost, which for a small business is not theoretical. If you commit $2,400 in January for a year of something, that is $2,400 you do not have in March when a client pays late. I have twice paid an invoice late — incurring an actual fee — while sitting on annual subscriptions that were, in cash terms, prepaid inventory I could not sell back. The twenty percent looked cheaper than the overdraft. It was not.

The second is the cancellation asymmetry. Read what happens if you leave mid-term. In most agreements I have looked at, the answer is nothing happens: you keep access until the term ends and you get no money back. This is not unreasonable and it is clearly stated. But it means the discount is contingent on a prediction — that you will still want this in month nine — and you are the one bearing the risk of that prediction being wrong. A twenty percent discount that you forfeit entirely by leaving in month four is not a twenty percent discount. It is a bet.

The third is what it does to your review habits. This is the one I did not see coming and now think is the biggest. A monthly subscription presents you with a decision twelve times a year. An annual one presents you with a decision once, and that decision arrives as an automatic renewal notice you will read in about four seconds. Over three years I have found two tools we were paying for annually and had not opened in months, and in both cases the reason we did not notice was that there was nothing to notice — no monthly line item, no recurring prompt, just one large charge that looked like a thing we had already decided.

So when is annual right? When the tool is genuinely load-bearing — proven by a week without it, not assumed — and has been for at least two renewal cycles, when the amount is small enough that the cash timing does not matter to you, and when you have some other mechanism — a calendar entry, a quarterly audit, something — that forces a real review at least once during the term. All three, not any one.

There is a version of the deal that is better than both and almost nobody asks for it: monthly billing at the annual rate, in exchange for a twelve-month commitment. Vendors will sometimes agree to this, particularly below a certain deal size where nobody is going to escalate it, because their actual concern is churn and not float. I have asked eleven times and been told yes four times. That is a worse hit rate than I would like, and it is four times I got the discount without making the loan.

The reason I write this down every time is that the annual toggle is the single most reliably taken default in software buying, and defaults get taken because they are defaults, not because they were evaluated. Twenty percent is a real number and sometimes the right answer is still yes. It is just worth being the kind of buyer who arrives at yes rather than starting there.