Owners and managers might intuitively know everything their guests are thinking about the food and service, but the data from reviews makes it definitive.
Most restaurants see online reviews as stars or star ratings. That's literally just the visible part of the review universe. There is so much more out there.
Review data's ability to help owners create a better guest experience is virtually infinite. It just takes a little analysis. For example, a prominent, iconic east coast seafood restaurant was struggling with more bad reviews than it should have been getting. By looking at the negative review data, three simple things that were bringing scores down jumped out:
Whenever "clams" was mentioned (24 times in the period analyzed), those reviews averaged 2.33 stars. Interestingly, every time "oysters" was mentioned (20 times), those reviews averaged 3.9 stars. The restaurant started suggesting oysters over clams, and the negative clams reviews were significantly reduced as the restaurant worked on sourcing better clams.
"Bread" was mentioned in 16 reviews which averaged 2.38 stars. In each case, the reviewer complained about no bread being served. The restaurant used to serve free bread on every table, but they saw that was creating a lot of waste–people weren't really eating the bread. So they stopped automatically serving it. But clearly, guests wanted the bread on the table and thought the restaurant was skimping by not serving it. As a compromise, the restaurant started having the servers ask guests if they wanted bread. This eliminated almost all those bad reviews moving forward.
"Wait" was mentioned 14 times, and those reviews averaged 1.4 stars (people were saying they waited up to 1.5 hours for their food to arrive). Of course, the restaurant knew they were having some service delays but weren't exactly aware of the damage it was creating.
Owners and managers might intuitively know everything their guests are thinking about the food and service, but the data from reviews makes it definitive; where you don't have to think it, you can see it.
Eliminating bad reviews by knowing the data is valuable not only helps you improve guest experience, it often takes your rating on review sites to the next level. For example, if a restaurant averages 3.74 in its ratings, it is displayed as 3.5 stars. To have a rating shown as four stars, they only need to average a 3.75, or 1/100th of a star change. Gaining that advantage is often a matter of making a few tweaks, such as what was mentioned above.
What's most important is that restaurants don't make reactions based on individual reviews. The input from a single, prominent review could be rogue and taking action because of it might cause more problems than it solves. You absolutely have to pay attention to it, but make changes based on the review data aggregate. A client of ours in Chicago, known for its pasta dishes, received an incendiary review on the saltiness of its Bolognese sauce. The guest said no Italian should ever be forced to suffer through that sauce and the subsequent water-drinking that follows. But we let the client know over the past six months, their Bolognese had a Net Promoter Score of 89.
Net Promoter Score is a customer satisfaction measurement that represents the percentage of promoters (5-star reviews) minus the percentage of detractors (1, 2, and 3-star reviews). A 4-star review is neutral and doesn't impact the rating. The scores range from -100 (all negative reviews) to +100 (all positive reviews). Any NPS score above 0 is considered "good," above 50 is considered "excellent," and above 70 is considered "world-class." In this case, there were 41 positive mentions of Bolognese and only five negative comments. Nothing needed to be "fixed" with the Bolognese.
Maybe you believe more in yourself or your chef than the guests? Read James Surowiecki's seminal book called The Wisdom of Crowds, Why the Many are Smarter Than The Few. There he proves that large groups of people are smarter than an elite few in regards to solving problems, figuring out solutions, and predicting the future. When you trust the data, associate a sentiment score to each keyword you care about in your reviews, you can start to gather some incredible insights and start to see that, too.