The Architecture of Disappearing: How Dating Apps Engineered Ghosting
Dating app design with infinite options and zero switching costs makes ghosting a rational choice, not a moral failing.
The psychology of choice overload—illustrated by Sheena Iyengar's jam experiment—explains why dating apps produce ghosting as an equilibrium outcome. When platforms offer unlimited swipes with no switching costs, investing emotionally in any single match becomes economically irrational. Features that provide instant social feedback and perpetual access to alternatives incentivize rapid, low-investment decisions. Ghosting persists not because users lack manners, but because the architecture of these platforms makes disappearing the path of least resistance. Changing the behavior requires changing the design.
In 2000, Sheena Iyengar’s grocery-store experiment showed that when shoppers faced 24 jams, they were less likely to buy than when they faced six. Too much choice, it turns out, makes people walk away. Dating apps take that dynamic and accelerate it: not 24 jams on a table, but hundreds of profiles per session, each swipe costing nothing, each match replaceable within minutes. Ghosting—ending a conversation without explanation—is often framed as a moral failure. But from a behavioral-design perspective, it’s the equilibrium outcome of a system engineered for perpetual evaluation. Research on platform affordances confirms that features like unlimited swiping and instant social feedback incentivize rapid, low-investment decisions. When every interaction is abundant and the next option is always a thumb-flick away, investing emotional energy in any single match becomes economically irrational. If you were designing a rational agent to navigate a market with near-zero switching costs and asymmetric information about partner quality, you’d program it to disappear the moment a better prospect appeared. The problem isn’t the agent; it’s the market design that rewards constant search and punishes commitment. Ghosting persists because the architecture of these platforms makes it the path of least resistance. Changing that behavior means changing the design that produces it.