Onboarding commitments front-load setup steps — profiles, preferences, imported data, configured devices — so that the effort a user invests early becomes a switching cost that keeps them from leaving later. The same steps that genuinely personalize a service also manufacture attachment: the more effort someone puts into building something, the more they value it, an effect documented as the “IKEA effect.” Combined with sunk-cost reasoning, that invested effort turns into a psychological leash long after the product’s ongoing value would justify staying.
The mechanism is legitimate and manipulative uses looking identical at the moment of setup. Real personalization and deliberate lock-in both ask for the same up-front investment; the difference shows only later, in the asymmetry between how easy it was to join and how hard it is to leave. When cancellation is buried, data cannot be exported, and the accumulated configuration is invoked to guilt a user into staying (“look how much you’ve set up”), the onboarding was designed as a trap rather than a convenience.
Because the tell is structural rather than in-the-moment, the defense is simple but must be applied early: inspect the exit before paying the entry cost. Locate the cancellation path and the export option before investing hours of setup, and when weighing whether to stay, value the service on its present merits alone — the effort already spent is gone regardless of what you decide next.