Professor of Cyber Security, Shujun Li, from the School of Computing and the Kent Interdisciplinary Research Centre in Cyber Security (KirCCS) has commented on the challenges of reputation systems to fully manage the authenticity of their users due to freedom of speech and provides advice for businesses on communicating productively with users who have left negative reviews.
Customer reviews are at the core of many real-world reputation systems. From sector independent online review services like TrustPilot and peer-to-peer (P2P) online shopping platforms like eBay, to online booking websites like Booking.com, potential customers and businesses rely on credible testimonials. Yet reviewers can sometimes receive backlash for their feedback from businesses, facing questioning over their authenticity.
Shujun says: ‘A healthy reputation system heavily relies on honest reviewers who provide genuine and accurate reviews, but (both negatively and positively) biased, inaccurate, fake and even malicious (defamatory) reviews have been a common phenomenon on many of such systems. In order to control quality of reviews, most reputation systems have a mechanism to verify if a reviewer is a genuine customer or consumer, but this does not always work due to reviewers’ privacy concerns and the conflict of interest of the reviewee. Even if a reviewer’s experience can be verified, there is often no way to verify the genuineness of the review since by definition it is supposed to be the reviewer’s personal opinion.
‘If businesses challenge customers who have produced negative reviews too aggressively, it can deter other honest customers from providing genuine (both negative and positive) reviews. It could also drive future unhappy customers to behave more anonymously and more negatively. Both are unwanted behaviours for the healthy running of a reputation system and will ultimately harm both customers and businesses. Therefore, I would suggest businesses affected by negative reviews find ways to communicate with unhappy reviewers and try to improve so future customers will leave more positive reviews. If a specific review does turn out to be false, negotiating with the online website to get the review challenged would be a more sensible route. Many online reputation systems, like TrustPilot for example, actually have established procedures and technical means for responding to reviews and flagging false reviews.
‘From the perspective of protecting reviewers, there are ways for a reviewer to avoid being identified by the reviewee. Many systems allow reviewers to remain anonymous and it is up to reviewers to allow themselves identified by the reviewee. Some reviewers may choose to use “false” accounts to cast (true or false) reviews, which is another issue deserving its own solution.’