By Alexandre Calapez
Try a simple test: open the app for a streaming service, a gym, or software you no longer use, and try to cancel it. Not search for how to cancel it. Actually cancel it, in less time than it took you to sign up. If you manage it in under two minutes without talking to anyone, you’re the exception, not the rule.
There’s a technical name for this: dark patterns. And the subscription industry has turned it into an art form.
An Industry Built on Other People’s Forgetfulness
What makes this scheme so effective is that each individual charge feels insignificant. Nobody loses sleep over $9.99 a month. But multiply that across several forgotten subscriptions, across millions of households, and the result is a billion-dollar industry built on people’s distraction, not on the actual value it delivers.
An international sweep conducted by consumer protection regulators, in coordination with the US FTC, reviewed hundreds of subscription services and found that 76% of them use at least one deceptive design technique to obstruct cancellation, and 67% deploy several at once. The most common methods are consistent across industries: burying the cancel button several layers deep in account settings, requiring a phone call during limited business hours, guilt-inducing messages designed to trigger hesitation (“Are you sure? You’ll lose ALL your benefits!”), and successive confirmation screens that test the patience of anyone who simply wants to leave.
This isn’t a portrait of imperfect design. It’s a portrait of an inverted conversion funnel: optimized, tested, and refined with the same engineering rigor any growth team applies to the sign-up process, just pointed in the opposite direction.
The Case That Should Have Changed Everything, And Still Hasn’t
In 2023, the US Federal Trade Commission (FTC) sued Amazon, alleging that the company used deliberately deceptive interfaces to enroll consumers in Prime without clear consent, and then “knowingly complicated” the cancellation process, to the point that a former FTC official publicly compared the cancellation journey to a multi-page, multi-option odyssey. In September 2025, Amazon agreed to pay $2.5 billion to close the case: $1 billion in civil penalties and $1.5 billion in refunds to roughly 35 million affected consumers. Without admitting wrongdoing, of course.
This should have been the definitive example of deterrence. Except the regulatory story around it is more revealing than the fine itself. In 2024, the FTC had finalized a federal rule, known as “Click-to-Cancel,” requiring companies to make cancellation as easy as sign-up. That rule was struck down in court in July 2025, after a legal challenge from industry. The agency has since recodified an earlier, narrower version of the rule, in effect since February 2026, and opened a new rulemaking process to try to restore a stronger version. In other words: more than two years after the most high-profile case on this issue became public, there still isn’t, in the US, a fully enforceable federal rule requiring symmetry between signing up and cancelling.
The Thesis: This Shouldn’t Be Treated as “Bad User Experience”
Here’s the central argument, and it’s deliberately uncomfortable: when a company keeps charging a customer for a service they’ve clearly and repeatedly tried to stop using, that isn’t a design flaw. It’s money retained through active deception, the same thing that, if done by a person secretly copying someone else’s credit card, we’d immediately call fraud.
The only difference is scale and form. Instead of one visible, one-off scam, it’s a $9.99 charge repeated quietly, tolerated precisely because each individual instance seems too small to justify legal action, and that’s exactly where the business logic lives. A company that knows, with precision, how many customers are paying monthly with no intention of continuing to use the service, and consciously chooses to make it harder for them to leave rather than easier, isn’t making a UX mistake. It’s choosing to profit from the confusion it designed itself.
The Counterpoint That Also Deserves Space
Not all friction in cancellation is malicious. There are legitimate reasons to ask for confirmation before ending a service: avoiding accidental cancellations, understanding why a customer is leaving, or offering a genuine discount to someone hesitating over price. Telling that apart from deliberate manipulation takes judgment, not just outrage: the dividing line is whether a company is informing a decision or obstructing one.
But the practical distinction is simple to apply, and it’s actually the basis of the very regulation courts have been blocking and recodifying: if the path to start paying takes one click, and the path to stop paying takes six screens, a mandatory phone call, and a representative trained to refuse the request three times before escalating it, that isn’t care for the customer. It’s a timed obstacle course built to squeeze out one more month of billing.
Calling it “aggressive business practice” is, once again, softening what it actually is: charging someone for something they’ve already made clear, in no uncertain terms, they don’t want.
