We kept every email announcing a price change for twelve months across a household’s ordinary subscriptions — streaming, storage, software, music, a couple of apps. Eleven arrived.
The numbers
| Services that raised prices | 11 |
| Median increase | 14% |
| Largest single increase | 29% |
| Services that raised twice in the year | 3 |
| Notices whose subject line said “price” | 0 |
The compounding is the part that surprised us. Three services raising prices twice in twelve months means an annual increase well above the headline figure of either notice.
What the notices are called
Not one of the eleven emails used the word “price” or “increase” in the subject line.
The actual subject lines included “An update to your plan”, “Changes to your membership”, “We’re making some improvements”, and — our favourite — “Your [service] is getting even better”.
That is not illegal and it is not accidental. A subject line that does not announce a price rise does not get opened, and an unopened notice becomes an unchallenged charge thirty days later.
The structural reason this works
Subscriptions renew by default. That is the entire mechanism.
A price rise requires no action from the company beyond a notice and no decision from you beyond inertia. Cancelling requires you to notice the email, decide, find the cancellation flow, and complete it — four steps against zero.
The asymmetry is the product. Everything else is a detail.
What actually worked
Three things, in descending order of effect.
A calendar reminder eleven months out for every annual subscription. This is the single most effective habit we found and it costs thirty seconds per subscription. Most unwanted renewals are not deceptive; they are forgotten.
A dedicated card or a spending-notification filter. Seeing the charge is the thing that prompts the review. Bundled into a large statement, a £9.99 line never gets looked at.
Actually cancelling one thing. Two of the eleven services offered a retained discount when we started the cancellation flow — in one case a 40% reduction for twelve months. That discount was available at any time and was never offered proactively.
We want to be careful here: we are not recommending you threaten to cancel as a tactic. We are reporting that the price you pay is frequently not the lowest price the company will accept, and the only way to discover that is to begin leaving.
What did not work
Emailing to ask for the old price. Three attempts, three polite refusals directing us to the cancellation flow, where two of the three then offered a discount.
The flow has the authority. Support does not.
The arithmetic worth running once
Add up what you pay annually for every subscription. Not monthly — annually, because monthly pricing is designed to feel small.
Then ask of each line whether you would buy it today at that annual price, knowing what you know. In our household that question ended four subscriptions, and the aggregate was more than we had estimated by a factor of about two.
Inés Okonkwo
Editor
Founded this because product writing had stopped distinguishing between a measurement and a press release. Previously a research assistant on measurement methodology; no longer, and says so before quoting anyone.
Elsewhere in the magazine
- The best free calorie counter apps in 2026, and the three different things 'free' is hiding
- The introductory price problem: what these subscriptions renew at, and why we left half this table empty
- What calorie counting apps actually cost per year, and the six-times spread nobody mentions
- What happens when you try to leave: cancellation and data export across nine subscriptions