? The Mayo Clinic: An information company |
| Sci Fi's Webisodes ?
September 04, 2006
Online reviews? How much will you pay me?
How many of you leave online reviews for products and services? How many of you read them? If you're like me, you review next to nothing, but religiously read what others write. I'll admit, I'm a freeloader, a leech. Reviewers are creating enormous value for e-merchants, and for us, and they do it for free. T.L. Pakii Pierce cites a Forrester report that 48% of online shoppers view customer reviews as critical. (ex: Brand Autopsy)
Fact is, the review system as it stands now is skewed. The only people we read are those who are proud of their opinions and eager to share them. Quieter types may have opinions that are just as valuable, but we never read them. At Yahoo and elsewhere, researchers are looking into new incentives to turn more of us into reviewers. The company is hiring behavioral economists to help create new markets around reviews and ratings, turning quiet gawkers into active participants.
Who knows how much the dynamics of online business (not to mention blogs) would change if what Richard Nixon used to call "the silent majority" raised its voice.
Using some of Surowiecki's WISDOM OF CROWDS thinking ... online ratings and reviews lack a "diversity of opinion." There probably aren't enough people rom all facets of life motivated to share their ratings and reviews online.
And to your point, these online ratings/reviews systems are skewed as we need to get the quieter to become louder and the louder to become quiter.
Posted by: johnmoore (from Brand Autopsy) at September 4, 2006 10:51 AM
Interesting, I have been wrestling with the very same problem, though for me it was a company offering to send me a $200 bean bag chair to "review" on my weblog. Hmmm... here's what I ended up writing:
Probably a good in-the-trenches supplement to the research you cite, and definitely good reading for the PR folk out there.
Posted by: Dave Taylor at September 4, 2006 11:53 AM
I would argue the point that we don't hear from "quiet types" on reviews. Look at Amazon book reviews. Many of the top reviewers read books constantly, offering insightful feedback. I would bet that many of these bookworms are not the type to grab your attention at the local bar. Quite frankly, I think that the open review system allows you to hear these softer voices that never would have been heard before. The focus is changing a bit from the well spoken, or those in media (journalists too) who make a living sharing their thoughts. Now in the comfort of your own home, you suddenly have a voice in a larger forum.
Posted by: Dan Blank at September 4, 2006 07:20 PM
Reviews can and sometimes are also be abused by vendors posting deliberately positive reviews through company employees.
Posted by: Harry at September 5, 2006 04:36 AM
Dan, I would agree with you that many of the Amazon reviewers are probably not loudmouths in other areas of their lives. But on Amazon, they strut their stuff. In effect, in that microcosm, they are loudmouths. And perhaps the guy who makes all the noise at the local bar or at Starbucks is quiet as a mouse on Amazon. The point, as John Moore notes, is that these systems would benefit from wider collaboration.
Posted by: steve baker at September 5, 2006 02:19 PM
I think product reviews should be reviewed before postings to eliminate the ALL CAPS worthless posts that are just there because of some annoyed customer for one reason or another. Will people then weed out reviews they don't like, maybe, but either way someone un-affiliated with companies should be checking these out so consumers can read things worth reading and not have to weed through 500 reviews to get 20 good ones.
Posted by: Tony Alaimo at September 5, 2006 04:47 PM
There's an interesting study done by MIT which included a section titled: "Do online reviews represent the opinions of the population at large?" Where they surveyed 1,970 people from a "nationally representative pool" and compared it to online ratings of movie reviews. The results were pretty close. You can see a snapshot of the graph on my blog http://blog.boorah.com/?p=16 and also link to a download of the report.
Posted by: Eric at September 5, 2006 07:37 PM
The primary reason I don't review is the many poorly written privacy policies of the review sites. Too often giving a review means giving an email address that is then used for marketing, etc. Even if it isn't, too often there is no guarantee or it would take a lawyer to interpret the policy.
Posted by: Richh at September 5, 2006 10:15 PM
It's an interesting issue - another side to the issue that I've recently been thinking about is whether the utility of these kinds of reviews (and the extent of user participation) varies across types of products. For example, are user reviews more useful if you're shopping for a blender than if you are thinking of buying a CD?
If anyone is interested in this issue with regard to music, I recently posted http://studioues.blogspot.com/2006/08/editorial-music-recommendation-in-long_30.html on the subject and am taking a poll.
With regard to what prompts people to write reviews or just read them, I personally think it has more to do with emotion than with whether the individual is quiet/shy or outgoing/fearless. Having a particularly emotional good or bad experience with a product is probably going to increase the likelihood of posting a review - so in that sense, I agree with you that the reviews are skewed (If you read a book and felt it was "fine", did you feel prompted to write up a review? Probably not - you probably forgot about it quickly and didn't even think about writing a review. But you would if it were horrible and didn't want people to waste their money or if it were unexpectedly great and wanted to give it props). Another possibility - having an emotional response to another user's comment/review can draw in lurkers. I rarely post, but if the discussion is lively or if I feel the need to defend the product against another reviewer's "unfair" comments, I'll jump in.
Posted by: Tricia at September 6, 2006 12:29 AM
I work for a company called Wize.com and we??e been working on a number of the issues that have been brought up in the comments above. Just as Richh pointed out, I??e noticed that people who write reviews usually have a definite opinion of whether or not a product is good or bad and this makes them appear to be somewhat overzealous at times. It seems to me that reviewers are motivated to write because of their emotions toward a product more so than their personal desire for attention. If this is true then it is likely that user reviews are coming from a more random sample of the population that their opinionated tone might lead us to believe. But is it true? The MIT study Tony Alaimo mentioned suggests it is, but it would be nice to see more research.
Posted by: David Rupert at October 4, 2006 05:09 PM
alaTEST.com (yes, I also work for them) is an international review aggregator that has collected over 500 000 reviews from 600 sources. We have spent more than 18 months on developing a mathimatical algorithm that includes most product quality factors and weighting in order to calculate daily updated rankings of "probably the best" products on the planet according to the experts. We integrate user reviews as well, but at a far lower weight as well as normalizing the ratings. Note, the average rating of about 300 000 reviews in our dbase is 4.2/5 = 84%... just to demonstrate how biased user reviews are.
Posted by: Arie at October 26, 2006 03:51 PM
Yes, users do tend to be biased, usually towards one extreme or the other. That's why at our site, eDistiller, (yes, another employee weighing in) we not only weight the effect of reviews, but we weight the reviews themselves. Reviews that are highly biased may end being ignored completely.
Posted by: Rick at December 1, 2006 03:17 PM