The subscription audit, done once and then kept

A subscription is spending with the decision removed. You said yes once, months or years ago, and every charge since has happened without you being asked again. That is precisely what a household ledger is built to notice, and precisely what most budgeting advice gets wrong about subscriptions: it treats them as a category to slash rather than as spending that has simply stopped being reviewed.

The audit below is not a purge. It is the recording half of kakeibo, aimed at the one kind of spending that was designed to avoid it.

What actually counts as a subscription

The obvious ones are streaming, music and cloud storage, all billed monthly and all easy to spot once you go looking. The ones that matter more are the ones that do not read as subscriptions at all: a gym membership taken out as an annual contract and billed once, a magazine or a box that arrives quarterly rather than monthly, an app that offered a free trial and quietly started billing a card nobody watches closely. All of it belongs on the same list, because the test is not how often it bills, it is whether it renews without you being asked.

What does not belong on the list is anything with a genuine decision attached to each payment: a loan repayment, an insurance premium you actively renew and could switch, a bill that varies with use like electricity or water. Those already sit in Survival and a subscription audit does not change where they live. The list here is only the spending that would keep billing indefinitely with nobody in the room.

The number that changes minds

Every recurring charge has a monthly price and an annual one, and they are the same amount of money read two different ways. The monthly figure is the one printed on the billing page, and it is built, whether anyone designed it this way or not, to look small. Twelve dollars a month barely registers as a decision. One hundred and forty four dollars a year is a different sentence, and it is the exact same charge.

Annualising does not make anything more expensive. It removes the one trick a subscription has, which is billing in an amount too small to argue with, one month at a time, forever.

List every one, at its annual cost

This takes longer than people expect on the first pass, because the point of a subscription is to be forgotten, and forgetting is working exactly as designed until you go looking.

1. Go through a real bank or card statement, three months back, and write down every charge that repeats. Do this from the statement, not from memory. Memory is where the ones worth finding are already hiding.
2. Write the amount as it actually bills, weekly, monthly or annual, then convert every row to the same yearly figure so the list can be read and compared in one pass.
3. Note what each one is, in a few words, next to the number. A row that says "9.99, streaming" six months from now tells you nothing about whether it is still worth the charge.
4. Add it up. The total is usually the second surprise, after the individual annual figures. A household running the four category split calculator can put that total against a real month's four category split and see straight away which envelope absorbed it and how much room is actually left.

Sort them, do not purge them

Once the list exists, the next decision is not "keep or cancel." It is which of the two envelopes each row actually belongs in, and that question does the sorting for you.

• Optional is spending you would not particularly miss: a streaming service you forgot you had, an app you tried once, a service that duplicates one you already pay for. These compete for a place in the month against takeaway, a round of drinks, an impulse purchase, the same as anything else that lands in that envelope. Most subscriptions worth cutting live here, not because subscriptions are bad but because this is where the forgotten ones settle.
• Culture is spending you would genuinely miss: a language course you actually open, an instrument app used every week, a membership tied to a place you actually go. The test the four categories guide sets out applies here without modification: would losing it change how the month felt, not just how it looked on a statement.

The point of sorting rather than purging is that a Culture subscription which passes that test deserves to be defended out loud, at its full annual price, rather than cancelled on principle because it renews automatically. Cutting every recurring charge and calling the result discipline is not the same as deciding what the money is actually for.

One household, worked through

Take a household that goes through a card statement and finds seven recurring charges: two streaming services at 15.99 and 9.99 a month, a gym membership at 45 a month, a cloud storage plan at 2.99 a month, a meditation app at 69.99 a year, a language course at 12.99 a month, and a subscription box neither person remembers signing up for at 24 a month.

Annualised, that is 191.88, 119.88, 540, 35.88, 69.99, 155.88 and 288, a total of 1,401.51 a year across seven rows that individually looked small enough to ignore. Sorted, the gym and the language course go into Culture because both get used weekly and both survive the test of being missed. The rest, both streaming services, the cloud storage, the meditation app and the subscription box, go into Optional, and on inspection three of the five have not been opened in months. Cancelling those three alone frees roughly 550 a year, which a household putting the same amount aside every month can see mapped out on saving 500 dollars in twelve months, next to a saved figure rather than a vague intention. Nothing here required cutting Culture, and nothing here required cutting all seven.

Make it a habit, not a purge

The audit above is only worth the effort once if the list quietly regrows the following year, which it will, because free trials convert without a decision and prices creep a dollar or two at a time in a way no single charge ever justifies noticing on its own.

The fix is the same one kakeibo applies everywhere else: a fixed date rather than good intentions. Pick a month, write it into the ledger the way a birthday or a renewal gets written in, and run the same three steps again. A household that only revisits its subscriptions when one of them suddenly feels expensive will miss the ones that stayed cheap and stayed unused, which are usually the larger share of the total.

This is also where the audit meets the rest of the method rather than sitting apart from it. Logging each subscription by hand on its billing day, the practice the credit card guide already recommends, is what keeps the list current between audits instead of stale until the next one. A subscription you personally re-enter twelve times a year has already been considered twelve times, and the annual audit only has to catch what slipped through.

Two people, one list

In a shared household, subscriptions are the easiest spending to lose track of, because they are the easiest to assume the other person is watching. Run the audit as one list rather than two, with both people going through the statement together, because a subscription taken out by one partner and forgotten by both is the most common row on this list, not the rare one. The same principle the couples guide argues for the whole ledger applies here at a smaller scale: one shared record, reviewed together, rather than each person half watching their own half of it.

Where the freed money goes

Money that stops leaving on a subscription does not need a new purpose invented for it. It goes exactly where any other freed Optional or Culture spending goes: to whichever the ledger already points at first, the savings promise or a debt above its minimum. Running the safe to spend today calculator right after an audit shows the difference immediately, because the same income now has fewer fixed exits before the day even starts. A household that wants to see what the freed total does to the wider month can put it through the savings rate calculator, which is a more honest comparison than guessing at what "cutting subscriptions" is supposed to be worth.

None of this needed a resolution or a productivity system. It needed the same thing the rest of kakeibo needs: writing the real number down, sorted honestly, on a date that does not depend on remembering to be annoyed.