Why People Do Not Cancel Subscriptions, According to Card Data
Liran Einav, Ben Klopack, and Neale Mahoney used data from a large payment card network covering roughly 30 percent of subscribers to ten services, and found cancellations jump in the months when a card is replaced and renewal requires an active choice. The average monthly drop in retention during the replacement window was 0.08, four times the 0.02 drop in other months. They estimate these cancellation frictions roughly double seller revenues.
The natural experiment
The design is the reason this study is worth reading. Whether your credit card expires has nothing to do with whether you still want a subscription. But when the card is replaced, the subscription has to be re-authorized, which means doing nothing now results in cancellation rather than renewal. The default flips. Einav, Klopack, and Mahoney published this in the American Economic Review in 2025, using comprehensive data from a large payment card network that covers approximately 30 percent of subscribers to the ten services they study. If people were canceling because they had reassessed the value, card replacement months would look like any other month. They do not.
How big the jump is
Across the services, the common pattern is a sharp drop in retention around card replacement. The authors report that the average monthly drop in retention is 0.08 during the two-month replacement window, four times larger than the 0.02 drop during other months, and they note this understates the difference because they cannot pin down the exact replacement month. The size varies a lot by service, small for some and much larger for others, and it is quantitatively similar whether the card is replaced at 6, 12, or 18 months after sign-up. They also checked that the drop is not just people moving the charge to another card, since overall account activity recovers quickly after replacement.
What the authors conclude from it
Using stylized models where inertia comes from either inattention or switching costs, they estimate that cancellation frictions roughly double seller revenues on average, holding initial subscribers fixed. Under the inattention model, inertia raises revenue by 87 percent on average, ranging from 14 percent for one service to more than 200 percent for another. They also test a policy counterfactual: requiring an active choice every six months would cut the overall revenue impact of inattention by about half. That counterfactual is the interesting one for a household, because it is the only lever in the paper an individual can copy on their own schedule.
What this does not say
It does not say a specific share of your subscriptions is wasted, and it does not say anyone is losing a particular dollar amount per year. Those figures circulate constantly and are not in this paper. The study covers ten services in one card network over a defined window, and estimates revenue effects for sellers, not welfare losses for consumers. Someone who keeps a service through card replacement might value it highly, and someone who cancels might have been about to cancel anyway. The defensible claim is narrower and still useful: a meaningful share of renewal happens because renewal is the default, not because a judgment was made.
The envelope change to make
Give subscriptions one capped envelope rather than letting each service sit in its own quiet corner of a bank statement. A single cap does the thing the card replacement did, on a schedule you control: adding a new service means visibly taking room from a service already in there, which forces the comparison the auto-renewal was designed to avoid. Set a calendar reminder to review that envelope once a year, and treat the review as an active choice in both directions, keeping and canceling both requiring a decision. The point is not to cancel everything. It is to make renewal something you did rather than something that happened.
Sources
Every source below was retrieved and checked. Findings are reported as the source states them.
-
Selling Subscriptions
— American Economic Review, 2025
Using comprehensive data from a large payment card network covering roughly 30 percent of subscribers to ten services, months during which cards are replaced and active renewal is required are associated with much higher rates of cancellation; the average monthly drop in retention is 0.08 during the replacement window versus 0.02 in other months; cancellation frictions are estimated to roughly double seller revenues on average, and the authors report that requiring an active choice every six months would reduce the overall revenue impact of inattention by about half.
Common questions
How much money do people waste on forgotten subscriptions?
Nobody can tell you honestly, and the widely quoted figures are survey estimates rather than measurements. This study does not produce a per-household waste number. It measures cancellations around card replacement and estimates the revenue effect for sellers, finding that cancellation frictions roughly double seller revenues on average. That is evidence that inattention sustains a real share of subscriptions. It is not evidence that you personally are wasting a specific amount, and you can find that out faster by listing yours than by reading an estimate.
Why is canceling so much harder than signing up?
The paper does not analyze cancellation flow design, so treat any explanation as inference rather than finding. What it does show is that the default matters enormously: when doing nothing means cancellation instead of renewal, many more people end up canceled. Any friction that keeps the default set to renew is worth revenue to the seller, and their models put that value at roughly double revenues on average. You do not need a theory of intent to act on that.
Does canceling and resubscribing later actually work?
As a budgeting tactic it does the same thing card replacement did, which is convert a passive renewal into an active decision. The study notes that about 20 percent of accounts that unsubscribe for two months or more return to the service, so resubscribing is a real and common path. The cost is your time and any promotional pricing you give up. That tradeoff is yours to make, and the research does not tell you which way to decide.
Run this budget on your phone
Envelope Budget puts these envelopes in your pocket. Assign every amount, log spending as it happens, and see what is actually left.
Get Envelope BudgetiPhone · manual entry, no bank connection · 7-day free trial