About 3 years ago, Yelp released numbers on its website detailing the breakdown by star rating of all of its reviews. Here is an article discussing the fact that most reviews averaged about 4 stars and providing a screen shot of Yelps rating distribution. This was the distribution for 2009:
I was reading through Matt McGee’s blog summarizing that the majority of online reviews are positive and I noticed that Yelp reviewers have gotten a lot meaner in the last few years. Here is Yelp’s current review distribution from their website:
1-Star Reviews on Yelp Up 71%
1 Star reviews increase from 7% of total reviews to 12% of total reviews in the last few years. This is an unwelcome trend for local businesses. To increase to 12% of their total review base, the number of 1 star reviews in just the last couple of years mathematically had to be much higher than 12%. Yikes!
Rants Not Reality
According to Yelp reviewers, more than 20% of consumers are dissatisfied (based on 1 or 2 star reviews on Yelp in the last 3 years and the trend line noted in the sentence above). Does that sound like reality to you? Do you feel like everyone is very unsatisfied with 1 out of 4 or 5 local purchases they make? No way. These are rants not reality that are exasserbated by a self-selection sampling bias.
The challenge with review systems has been creating a system that reflects reality. The truth is that the vast majority of local business customers are happy. When we have done randomized and blind phone calls to local business customers, we have consistently found that less than 10% of customers are dissatisfied.
The solution to the problem of a small sample size that does not accurately reflect the total group is to expand the sample size – i.e. get more reviews. How should a business get reviews? Ask for reviews and enable every format that potential reviewers want to supply the content back to you: customer calls, handwritten reviews, email invitations, social-enabled invitations and collection, web forms, etc.