Posted on

Cancellation journeys that are substantially harder than enrollment

The fastest way to understand a subscription interface is sometimes to compare two journeys.

How many clicks does it take to join?

How many does it take to leave?

If enrollment requires one bright button while cancellation requires account menus, hidden links, retention pages, phone calls, limited support hours, and repeated attempts to keep the service, the difference is not merely cosmetic.

It changes how much effort the customer must spend to stop paying.

The Federal Trade Commission’s dark-pattern report calls these roadblocks to cancellation. Examples include allowing people to sign up online but requiring a different channel to cancel, hiding the cancellation phone number, limiting the hours of a cancellation line, or forcing the user through repeated sales pitches.

See the FTC’s Bringing Dark Patterns to Light.

Friction becomes part of the business model

Every additional step creates another place where someone can give up.

The customer gets distracted. The phone queue is too long. The cancellation link is not where expected. A retention offer looks like the final confirmation but actually keeps the subscription alive.

The FTC’s case against Amazon over Prime alleges that Amazon made enrollment easy while maintaining a complicated cancellation flow internally called “Iliad.” The agency says consumers had to navigate multiple pages and offers before completing cancellation.

See the FTC’s Amazon Prime case.

Leaving should be a normal product function

A clear exit process does not need to be identical to signup in every detail. A business may need to confirm identity, explain when service ends, or warn about data loss.

But the purpose of those steps should be completing cancellation safely, not exhausting the user until another billing cycle arrives.

FTC guidance on subscriptions has repeatedly emphasized that cancellation should be easy and simple, and at least as easy to use as the method used to enroll.

See the FTC’s subscription dark-pattern guidance.

The clean test is practical.

Can an ordinary customer who knows they want to leave find the exit, understand what it does, and complete it without being rerouted through unrelated obstacles?

If joining feels like opening a door and leaving feels like escaping a hedge maze, the imbalance is the feature.

Posted on

Network effects that keep dissatisfied users from leaving

A social network is not valuable only because its software is good.

It is valuable because your people are already there.

That makes leaving a different decision from replacing a calculator app.

A better calculator can win one user at a time.

A better social platform may need to win your family, coworkers, hobby group, local businesses, school parents, favorite creators, and everybody else you actually came to talk to.

That is a network effect.

The crowd becomes part of the product

The UK’s Competition and Markets Authority described this problem directly in its 2020 study of online platforms and digital advertising. It found strong network effects in social media and noted that Facebook had by far the largest network. Other services were often used alongside Facebook rather than as complete substitutes for it. See the CMA’s market-study presentation.

That explains an otherwise strange behavior.

A person can complain about a service every week and still open it every morning.

The contradiction disappears once the user is not choosing only between products.

They are choosing between staying where the network exists and leaving without the network.

Dissatisfaction does not automatically create an exit

Suppose a messaging app makes an unpopular policy change.

One user can install a rival in thirty seconds.

That does not move the family chat.

If Grandma, the soccer team, three clients, and the neighborhood group remain on the old platform, the dissatisfied user may end up maintaining both apps. The rival becomes an addition instead of a replacement.

The European Union’s Digital Markets Act explicitly recognizes this problem for messaging. Its interoperability rules say large messaging gatekeepers benefit from strong network effects and require certain services to interoperate with competitors. The European Commission says the first third-party messaging services interoperable with WhatsApp launched in November 2025. See the Commission’s messaging interoperability overview.

That is an important structural change.

Instead of requiring everybody to migrate together, interoperability can let users on different services continue communicating.

The alternative has to preserve the relationship, not merely the feature

A competing platform can have:

  • cleaner software,
  • fewer ads,
  • better moderation,
  • stronger privacy,
  • lower prices,
  • and a nicer logo.

None of that solves the network problem if the people a user needs remain somewhere else.

This is one reason platform decline can continue longer than ordinary product logic suggests.

A company does not necessarily lose users the moment it makes the service worse.

The accumulated network can absorb some deterioration.

That does not prove every dominant platform is deliberately abusing lock-in.

It does explain why user dissatisfaction and user departure are different measurements.

In the Enshittification Machine, the crowd is not merely using the platform.

The crowd is part of what makes the platform hard to escape.