What it is
Loss aversion describes choices in which a loss feels larger than a gain of the same size. It is a central part of prospect theory (Kahneman and Tversky, 1979).
How strong is the evidence?
Contested. Loss aversion appears in many lab and field studies, but a critical review argues that the evidence for a general principle is weaker than often claimed and depends on context (Gal and Rucker, 2018). Responses to that review defend loss aversion in many settings. The practical takeaway for CRO: treat it as a hypothesis generator, not a law that guarantees a lift.
Why it matters for conversion
The strongest, least controversial application is avoiding surprise losses. Unexpected extra costs at checkout are the most cited abandonment reason among non-browsing shoppers in Baymard’s survey (Baymard). A fee that appears late feels like something being taken away.
Ecommerce example
Fictional. Showing delivery cost and date on the product page removes a late “loss” at checkout. See the shipping-surprise case study.
SaaS example
Fictional. At trial end, saying “Your projects and settings stay saved if you upgrade within 30 days” is true and reassuring. Saying “You’ll lose everything tonight!” when data is kept is deception.
Ethical boundaries
- Don’t invent losses: fake expiring discounts, resetting countdowns, false “last chance”.
- Don’t exaggerate losses to block cancellation.
- Do reveal real costs and commitments early, and describe real consequences plainly.
How to test it
Hypothesis pattern: removing or pre-announcing a cost the user currently discovers late will reduce abandonment at that step. Primary metric: completed orders or subscriptions; guardrails: refunds, complaints, average order value.