Why cancelling subscriptions feels harder than not signing up

Cost Me Research Desk · May 27, 2026

TL;DR. Signing up costs nothing psychologically. Cancelling feels like a loss — and losses register about twice as strongly as equivalent gains. That gap is why a $14.99 subscription you would never sign up for again is the same $14.99 you cannot bring yourself to cancel.

Pick a subscription you wouldn't sign up for if you weren't already paying. There's almost always one. The streaming service you stopped watching. The meditation app you used twice. The premium tier you upgraded to during a sale and now use the free features of.

The interesting question isn't why you signed up. The interesting question is why, having clearly answered “not worth it” for months, you still haven't cancelled. The answer is in three different pieces of behavioural research, and they all point the same direction.

The endowment effect

Kahneman, Knetsch, and Thaler (1990) ran a now-classic experiment with coffee mugs. Half the subjects in a classroom were randomly given a mug; the other half were not. They were then asked, the owners how much they'd sell for, the non-owners how much they'd pay to buy.¹

Under standard economic theory, the median sell and buy prices should match — the mug is the same object. In practice, owners demanded roughly twice what non-owners offered. Mere ownership, for a few minutes, doubled the perceived value of the mug.

The phenomenon is called the endowment effect, and it generalizes well beyond mugs. The moment you consider something yours, giving it up registers as a loss — which the brain weighs much more heavily than the equivalent gain of getting cash back. Your Spotify Premium subscription is yours. The $11.99 you'd save by cancelling is hypothetical.

The moment you consider something yours, giving it up registers as a loss. Money saved is hypothetical; ownership feels real.

Status quo bias

Samuelson and Zeckhauser (1988) showed, in a series of hypothetical-decision experiments, that people systematically prefer whichever option is presented as the default — even when they have no rational reason to. Their term for it was status quo bias.²

Their participants made choices about jobs, investments, and policy. When an option was labeled as the existing state, that option was chosen disproportionately often — even by participants who, presented with the same options without the framing, would have chosen differently.

Recurring subscriptions are a status-quo machine. The default is to keep paying. Doing nothing is the action; cancelling requires effort. And the effort is usually deliberately disproportionate — buried cancel flows, retention offers, “are you sure?” screens. The system is designed so that the path of least resistance is to remain enrolled.

The power of defaults

Johnson and Goldstein (2003) showed exactly how load-bearing defaults can be. They studied organ-donor consent rates in European countries that differed only in whether the default was opt-in or opt-out (Johnson & Goldstein, 2003).³ In opt-in countries, consent rates ran 4–28%. In opt-out countries, 86–100%.

The decision is the same. The values of the population are presumably similar. The only difference is which choice you have to actively make. That single design choice — what counts as “doing nothing” — moved organ-donor consent by sixty to ninety percentage points.

Subscriptions exploit the same lever. The default is recurring billing. Cancelling is the active choice. And the asymmetry compounds because the cost of the wrong default falls on you, not the company.

Why the cancel flow is built the way it is

Companies that sell subscriptions are aware of all of this research. Many have data scientists optimizing cancel flows specifically against churn metrics. The result is a well-documented design pattern:

• Cancellation is hidden several layers deep in account settings, while sign-up is one button on every page.
• Retention offers — “wait, here's 50% off for six months” — activate the endowment effect one more time before you can leave.
• The cancel button is often labeled “Continue” rather than “Cancel,” or paired with a larger, brighter “Keep my subscription” button.
• Some flows require phone calls, support tickets, or confirmation emails — forms of friction that don't exist at sign-up.

None of this is illegal. Most of it is now regulated in some markets — the FTC's click-to-cancel rule and the EU's consumer-protection updates both target this exact asymmetry. But the underlying psychological gradient remains, and even well-designed flows can't eliminate the endowment effect.

The numbers

The average US household pays for roughly 9–12 active subscriptions across streaming, software, fitness, and utility services, according to multiple consumer-finance surveys in recent years. The median dollar amount is commonly reported in the $200–300 per month range, and the same surveys consistently find that households materially underestimate their own total — often by 2x — when asked to guess before being shown their actual statements.

$250 a month, compounded at 7% real for 30 years, is roughly $235,000 of foregone investment. That's the cost of subscriptions you would mostly cancel if you had to re-sign-up to keep them.

A simple intervention

The cleanest behavioural fix to the asymmetry is to invert it. Once a year — calendar it — go through every recurring charge on your card statement and ask: if I weren't already paying for this, would I sign up today at this price? If the honest answer is no, cancel.

That single question disarms the endowment effect by forcing you to evaluate from the non-owner's perspective. It also turns status-quo bias around — the question reframes “cancel” as “decline to sign up,” which is the action you would take by default if presented with the offer fresh.

Most people running this exercise the first time cut 20–40% of their recurring spending. The savings tend to stick, because the things you cancelled don't re-acquire status-quo protection — they have to win you back as new customers.

The honest limitation

Not every subscription is a behavioural-economics trap. Some genuinely earn their keep — and the endowment effect is the same psychology that helps you stick with good habits, too. The argument is not that all subscriptions are bad. It's that the asymmetry between signing up and cancelling means your subscription mix will, without intervention, drift steadily in the wrong direction.

What this means for you

Put a recurring calendar event on your phone right now — first Saturday of every January — labeled subscription audit. Forty-five minutes once a year is enough to reverse the gradient. Anything you actively re-sign-up for keeps the endowment-effect protection it's earned. Anything you wouldn't re-sign-up for at that price is, by definition, a purchase you wouldn't make today — and the only reason you're still making it is that you haven't had to make it.

References

1. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98(6), 1325–1348. https://doi.org/10.1086/261737
2. Samuelson, W., & Zeckhauser, R. (1988). Status quo bias in decision making. Journal of Risk and Uncertainty, 1(1), 7–59. https://doi.org/10.1007/BF00055564
3. Johnson, E. J., & Goldstein, D. G. (2003). Do defaults save lives? Science, 302(5649), 1338–1339. https://doi.org/10.1126/science.1091721

Want more like this? Loss aversion and the things we keep or cutting $80/month from streaming. Or head back to costme.io.