Commentary

...Or Your Lying Eyes?

Sockpuppeting -- using a false name to post online comments or reviews -- is one of the more annoying aspects of living the Web-centered life. Not long ago, some well-known authors were outed for having tried to up the popularity of their books on Amazon by writing fake positive reviews for their own books and posting damning reviews of authors considered their competition.

Meanwhile, a study of consumers who had recently made an electronics purchase found that shoppers were more likely to pay attention to consumer reviews than professional reviews across every product category. Consider this in the context of a claim by Ed Keller that word-of-mouth -- conversations at Super Bowl parties, at the water cooler the next day and the days after that, as well as social media -- is a better indicator of which Super Bowl ads generated the most buzz rather than the usual polls, or just Twitter and Facebook activity (the fact that Ed's firm specializes in word-of-mouth marketing notwithstanding.)

All of this is a long way around to lamenting that it is getting harder to know who and what to trust when researching online. Like the folks in the electronics purchase survey, I tend to read a fair number of consumer reviews before booking a hotel, restaurant or resort unknown to me and thanks to some hard lessons learned about buying impulsively online, I also increasingly look for consumer insights before making nearly every online buy.

Just like there are now a dozen ways to measure who wins the Super Bowl ad war each year, there are too many unknown paths to trustworthy opinions. Should I rely on Facebooks likes? The percentage of positive tweets? Professional reviews from trusted media brands? Consumer-generated reviews that may or not have been ginned up by someone with an interest in the success of the "product" I am considering? I know many social-media-based sites claim to have methods to ferret out bogus reviews, but I am not sure I trust them, either.

Word of mouth can be tricky as well. Just ask five of your friends what they thought of “Zero Dark Thirty” or “Silver Linings Playbook” and see if you get a consensus. How many times has one of your friends loved a movie that you saw on their recommendation, and you just hated it? Since we tend to read similar books, my friend Gaffney and I trade opinions on books all the time, yet there have been times we have disagreed on what was a good read and what wasn't.

So far I have found recommendation engines to be ineffective. I must say that Amazon does as good a job as any, but I can't recall having moved ahead on any of their ideas based on my past purchases of about 25 books a year. But I appreciate the effort. I keep a close eye on Flixster's Rotten Tomato index to see how critics and viewers rate movies they have seen -- and since their sample base is so gigantic I tend to think it is hard to influence it and that a concerted effort to influence it would be easily discovered and discounted.

But I can't help but think that nearly every other ranking or poll is gamed in some way. People routinely tell pollsters what they think the survey sponsor wants to hear, and if hacker types are skilled enough to scam Google's pay-per-click model (and you know they do), you have to assume they can make the worst resort in the Caribbean seem like a veritable paradise.

So, billion of transactions, comments, cries into the social media void and algorithms later, are we any better off?

Recommend (3)