The alert comes through on a Tuesday morning while you are still in the car park. A charge for $47.99, from a company whose name you half recognize and cannot place. You sit there for a minute going through it. Something to do with the printer? A thing your partner signed up for? By the time you are inside you have decided to look into it later, and later does not arrive.
What stops this is not an app that watches your account. It is one written list, with a date on every line, that you read on a day you have picked. Every charge you have ever been surprised by was on a page somewhere, agreed to by you, on a date you could have known. The surprise is a memory problem, so the fix is a written one.
Why a charge you agreed to still counts as a surprise
You agree once, in about eleven seconds, at the moment you most want the thing. Then the agreement goes quiet. A monthly charge is small enough that it never triggers a second look, and a yearly one goes so long between appearances that by the time it comes back around, the version of you that signed up has moved. Nothing has gone wrong. The charge is just arriving from further away than your memory reaches.
Start from the statement, not from memory
Sit down with the last three months of every card and account you use, and read every line. Three months catches the monthlies. It does not catch the yearlies, so add the ones you can remember and expect to add two more over the next year as they surface. Working from memory alone gets you about two thirds of the list and the missing third is where the surprises live.
This is the part that costs you something real. An hour, maybe an evening if the statements are spread across two banks and an app store. It is the only hour in this that is genuinely work.
What every line needs on it
- What it is, in the words you would use out loud, not the name on the statement.
- What it costs, and how often. Monthly, yearly, quarterly, weekly.
- The date it next renews.
- How it gets paid. The name of the card or the account, like Visa or checking. Never the number.
- Whether it is one you want to keep.
- Where you cancel it, which is usually a link, and which is the single most useful thing on the line.
The cost column only works if everything is comparable, so convert each one to a monthly figure as you go. A quarterly charge divided by three, a yearly one divided by twelve. Otherwise you are looking at a column of numbers that cannot be added up.
What to do with the ones you want to stop
Mark them rather than cancelling them on the spot. Cancelling takes a login you do not have to hand, and a half-cancelled list is worse than an honest one. Put the date you have to act by next to each, which for most subscriptions is the day before it renews, and do them in one sitting. Then keep the line on the list with its status changed, so that next month you can see it is done rather than wondering whether you did it.
If you would rather not build the columns yourself, we make a file that already has them and works out the monthly and yearly figures as you type. It is worth knowing that trials run on a different deadline to the one you would expect, and that deadline is the subject of its own piece.
Read it once a month, on a day you pick
Pick the day after your usual payday and put it in the calendar as a repeating event with the list attached. Five minutes. You are looking for two things only: what renews in the next thirty days, and whether anything on it has stopped being worth the money. That is the whole review.
How this actually fails
Not abandonment. Staleness. The list is built properly, it is right for about six weeks, and then three things sign up that never make it on because they happened on a phone at eleven at night. By month four it is right about most of what you pay for, which is the same as being wrong, because you have stopped checking the parts you assume it covers.
The way through is to add the line at signup rather than at review. Thirty seconds, at the moment you are already thinking about it, and the monthly read stays a read instead of turning back into an audit.