The arithmetic nobody performs
£12.99 is a rounding error. £11.99 is nothing. £8.99 barely registers. Subscription pricing is engineered around precisely that reaction: each charge is individually trivial, which is exactly why nobody adds them up. Eight of those, averaging £13, is £1,248 a year. Not a rounding error. For many households it exceeds what they put into a pension, and unlike the pension it renews silently whether or not anyone still uses it.
None of this is an argument for frugality theatre. Plenty of subscriptions are excellent value and you should keep them; the point is that you cannot judge value on a number you have never seen. This page makes you look at the annual figure, once, deliberately.
Building the list
Start with every charge that repeats: streaming, music, cloud storage, gym, software, news, delivery memberships, mobile add-ons, insurance paid monthly, apps you subscribed to inside your phone and forgot. Set each one to bill monthly or yearly and the table converts everything to a comparable per-year figure: monthly charges are multiplied by twelve, annual ones pass through as they are, and the monthly equivalent sits beneath the total. That is right for totalling, but an annual bill's cash still leaves in one lump, a timing problem belonging to the sinking funds planner.
The best source for a complete list is your bank statement, not memory. Scan three months and search for the same merchant appearing repeatedly; if you already keep an expense diary, the recurring charges in it are the fastest place to start, since it can spot the repeats for you. Check your app store subscription list separately too, because those charges often appear under the store's name rather than the service's, which makes them nearly invisible on a statement.
Whatever you find stays in your browser. The list is not sent to us and there is no bank connection, you type what you find, and that is deliberate: an audit tool should not require handing over read access to your account. It will still be there next month as long as you use the same browser and do not clear site data, and account sync so the list follows you across devices is planned.
Then use the keep checkbox as a decision, not a description. Untick anything you would not sign up for again today at that price, and the panel totals what those cancellations are worth. Judge each line on its own merits while you do it. A gym membership you use three times a week is outstanding value at £34 a month, and a streaming service that keeps the household happy is not waste; the annual total is a useful shock, not a savings target, and treating it as an indictment leads to a burst of cancellations, several months of resentment, and re-subscribing to half of them. Cancelling is not always free either. Some contracts have minimum terms, exit fees, or lose you a legacy price you can never get back, so check before you act on a number this page shows you. Working out what can go without losing anything you actually value is the harder half of this, and the audit guide has more on where to draw that line.
A negative investment
The final panel answers the question that actually motivates cancelling. A cancelled subscription is not a one-off saving of £13; it is a permanent monthly sum you can redirect. Put £39 a month into investments returning 7%, a rate worth sense-checking against the historical record for equities, and after ten years it is roughly £6,700. That figure is calculated the same way as the compound interest calculator, and if the money would be better used against a mortgage instead, the mortgage overpayment calculator will show the interest saved and the years knocked off the term. The link runs both ways: a subscription is a negative investment, compounding against you at exactly the same rate, in the same direction, for the same number of years.
That £6,700 carries two honest caveats. The projection assumes something behavioural that frequently does not happen, namely that the cancelled amount is genuinely redirected into investing rather than absorbed by ordinary spending. Money not spent on Netflix does not automatically become money invested; set up the standing order the same day you cancel, or the figure stays hypothetical. And the growth uses a fixed annual return applied smoothly, when real markets do not deliver 7% every year in an orderly line, and the number is in nominal terms unless you enter a real return. For how much that distinction matters over decades, see the inflation calculator.
The creep
Services rarely launch at the price you are paying now. They launch cheap, build the habit, and raise the price in small increments announced by an email you did not open. A streaming service at £5.99 in 2019 is commonly £12.99 or more today: the same product, at double the cost, without a single moment where you consciously agreed to double your spending. Run that £5.99 through the Bank of England's inflation calculator and you can see how much of the rise general prices explain, and how much is the service quietly repricing you.
This is why the audit is a monthly habit rather than a one-off spring clean. Re-open it, check each amount against your latest statement, and update anything that has moved; the pattern only becomes visible when you look repeatedly. Creep is also the strongest argument for monthly billing when you are unsure about a service: annual is usually cheaper per month, but it commits you for a year and makes rises harder to notice, while monthly buys you the option to leave.
Free trials deserve their own attention. A trial you forget is a subscription you chose by inaction. When you start one, put a reminder for two days before it converts, and add it to this list immediately with the post-trial price, not zero.