Forgotten subscriptions: the science of why your brain misses them

Open your last bank statement and circle every merchant you can't immediately identify. For most people, there are two or three. Some are legitimate purchases you've simply forgotten — a Tuesday lunch on the road, an Etsy order you never mentioned to your partner. But at least one will be a recurring charge: a service you signed up for once, a free trial that converted while you weren't paying attention, an app you opened twice in 2023 and never again. You're paying for it still. You haven't thought about it in months.

This isn't a memory failure. It's the predictable output of a system designed, deliberately or not, to bypass the cognitive machinery that normally tracks money leaving your account. Two strands of consumer-psychology research, both more than twenty years old, explain exactly how the mechanism works.

The first comes from a 2001 paper in the Journal of Consumer Research by Dilip Soman, then a professor of marketing at Hong Kong UST. Soman wanted to understand a question retailers had been asking for years: why do credit-card customers spend more freely than cash customers, even when the dollar amount is identical? His hypothesis was elegant. The act of paying with cash or a check forces the consumer to do something the act of swiping a card doesn't: rehearse the amount. You count the bills. You write the figure on the line. You sign your name beneath it. Each of those small motor and cognitive acts plants the spend in memory.

Soman called it the rehearsal hypothesis, and he tested it across a series of controlled experiments. The results were unambiguous. Subjects who paid by check recalled their past spending with significantly higher accuracy than subjects who paid by card. More strikingly, the high-rehearsal group also made smaller subsequent purchases, as if the memory of the prior outflow was actively constraining the next one.

Past payments strongly reduce purchase intention when the payment mechanism requires the consumer to write down the amount paid (rehearsal) and when the consumer's wealth is depleted immediately rather than with a delay (immediacy). — Soman (2001), Journal of Consumer Research

Translate this finding into the experience of a modern subscription. There is no rehearsal. There is no writing. There is no immediacy. The charge fires while you sleep, against a card you authorized months ago, in an amount you may never have explicitly approved. Both of Soman's mechanisms are at zero. The system is, by his own model's prediction, the most forgettable form of spending the human brain encounters.

The second strand of research deepens the picture. In a 1998 paper in Marketing Science, the behavioral economists Drazen Prelec and George Loewenstein introduced the concept of payment-consumption coupling. Their argument was that hedonic experience and financial cost are normally yoked together. You eat the meal, then you pay for it, and the two events form a single memory. Credit and subscription billing decouple them. You enjoy the service in May; you pay in June against a charge you barely notice; the two events never cognitively meet.

Decoupling isn't accidental. It's the explicit appeal of credit and subscription billing — the friction is removed from the moment of use, which is exactly what makes both products feel good. But the same friction is what would have stopped you, three months ago, from continuing to pay for a service you no longer use.

Credit mechanisms decouple the act of consumption from the act of payment… therefore weaken self-regulatory processes, enabling more spontaneous consumption. — Prelec & Loewenstein (1998), Marketing Science

The subscription, viewed through this combined lens, is a particularly hostile environment for the part of your brain responsible for noticing spending. Soman's rehearsal mechanism is suppressed because no human action accompanies the payment. Prelec and Loewenstein's coupling mechanism is suppressed because the use and the payment occupy entirely separate weeks. Both findings are now more than twenty years old. Both have been replicated. The mechanism is robust.

Knowing the mechanism suggests the intervention. If automatic billing erases rehearsal, you can restore rehearsal manually. Pull a 90-day window of your card statements once a quarter. List every recurring merchant. Write the amounts down by hand if you can stomach the friction; type them if you can't. The act of the list is the intervention. Soman's research suggests, fairly directly, that this single hour of effort restores the spending memory the system erases the rest of the time.

What the research doesn't say — but what the experience of actually running this audit consistently produces — is that the surprise on the first pass is rarely the size of any individual charge. The surprise is the sum. Each $9.99 looked reasonable in isolation. Together, they read differently. The same money, presented as a single decision, becomes a decision you'd actually make.

This is why subscription audits work. They don't catch new information. They restore old information your brain was systematically prevented from forming.