In just six years, Yelp.com has managed to crowdsource 20 million reviews of restaurants and other services by creating and leveraging an impressive social network of people who enjoy writing reviews. But can a bunch of amateur opinionators working for free really transform the restaurant industry, where heavily marketed chains and highly regarded professional critics have long had a stronghold?
To answer this question, HBS professor Michael Luca combined Yelp reviews with revenues for every restaurant that operated in Seattle, WA at any point between 2003 and 2009. Applying a new method to tease out the causal effect of reviews (separate from the effect of underlying quality), the study shows that a one-star increase on Yelp leads to a 5 to 9 percent increase in revenue.
Yet Yelp doesn't work for all restaurants. Chain restaurants —which already spend heavily on branding —are unaffected by changes in their Yelp ratings. This suggests that consumer reviews present a new way of learning in the Internet age, and are fast becoming a substitute for traditional forms of reputation.
This certainly confirms my experience. I use Yelp a lot -- mostly on the phone, to get map data and contact information. But I notice the stars. And I'm sure they have a subtle effect on my choices.
This seems like a very good thing. If a good reputation online can drive customers into a business, then maintaining that reputation will be as important as, say, advertising in a more traditional way. And, naturally, there will be an attempt by some businesses to game the system with phony reviews and paid-for recommendations. But Yelp may self-correct for that kind of thing:
...consumers respond more strongly when a rating contains more information. Consumer response to a restaurant's average rating is affected by the number of reviews and whether the reviewers are certified as "elite" by Yelp, but is unaffected by the size of the reviewers' Yelp friends network.
I'm a total junkie for this kind of research, especially when it involves web-based behavior. This really is a new phenomenon -- the ability to source and crunch so much data, so quickly, and with such granularity. We're all adjusting our decision-making processes to each new waterfall of data. I'm not sure I know what an "elite" status reviewer is on Yelp, but I'm certain that over the next year or so, I'll be influenced by several of them.