Wow. Recently there have been a number of news articles about the problem of fake reviews at rating sites. Most of these (here’s one example) were spurred by a report that the NY attorney’s general office set up a sting operation and caught 19 companies either posting fake reviews or offering to post fake reviews for a fictional yogurt store they set up.
While fake reviews are a real and important problem, they are just the tip of the iceberg. Probably a bigger problem are “real reviews” left by real customers – but ones that have been cherry-picked by the business by encouraging customers to post a review after the business knows they have had a good experience.
Another potential problem is the possible lack of representativeness of people that use reviews sites. To look at this issue we analyzed the demographics of the respondents to our 2013 Online Review Study. The full report can be found here. I think we found two interesting things. First, younger and more affluent respondents tend to use review sites at higher rates than others. We also know that these groups of individuals are “harder graders” on customer experience surveys in general. Second, these same groups also tended to trust the information less. That’s right, people that use the information most, trust it less. Why would that be?
From people’s comments it seems that while a lot of people think much of the information is suspect, they are confident in their ability to weed out the good from the bad. If they can really do that, that’s great, but I see one big problem here. It is a well-documented phenomenon in psychology called “illusory superiority.” If you ask people to rate themselves on pretty much any positive attribute, way more than ½ will rate themselves above average. You probably think you are an above average driver. So do about 90% of your peers. What does this have to do with evaluating postings on ratings sites? People probably aren’t as good at it as they think they are and they are getting “faked out” without knowing it.
By the way, I’ll be talking about these findings at The Market Research Event conference in Nashville on October 23. Hope to see you there.
As always, comments and thoughts about this post are welcome and encouraged!
