Why trust seals don't work: a study of user perceptions and behavior

  • Authors:
  • Iacovos Kirlappos;M. Angela Sasse;Nigel Harvey

  • Affiliations:
  • Department of Computer Science, University College London, London, United Kingdom;Department of Computer Science, University College London, London, United Kingdom;Department of Psychology, University College London, London, United Kingdom

  • Venue:
  • TRUST'12 Proceedings of the 5th international conference on Trust and Trustworthy Computing
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Trust seals, such as the VeriSign and TRUSTe logos, are widely used to indicate a website is reputable. But how much protection do they offer to online shoppers? We conducted a study in which 60 experienced online shoppers rated 6 websites --- with and without trust seals - based on how trustworthy they perceived them to be. Eye tracking data reveals that 38% of participants failed to notice any of the trust seals present. When seals were noticed, the ratings assigned to each website were significantly higher than for the same website without a seal, but qualitative analysis of the interview data revealed significant misconceptions of their meaning (e.g. "presence of seals automatically legitimizes any website"). Participants tended to rely on self-developed --- but inaccurate --- heuristics for assessing trustworthiness (e.g. perceived investment in website development, or references to other recognizable entities). We conclude that trust seals currently do not offer effective protection against scam websites; and suggest that other mechanisms --- such as automatic verification of authenticity are required to support consumers' trust decisions.