Why You Can Never Truly Trust Online Restaurant Ratings
As a veteran food blogger, I’m often asked about the importance and reliability of online restaurant ratings. In Lebanon’s F&B landscape, Zomato has filled the role of restaurant search and discovery. Users can look up business hours, menus, photos, diners’ reviews, and contact info to learn more about a venue of interest.
Online data can certainly go a long way in creating a virtual restaurant visit even before the real one transpires. And I was quoted saying that a couple years ago in an interview with Wamda on “How Lebanon’s food scandal spurred startup solutions.” But while one can piece together an impression from the general attitude of previous diners, one would be sorely amiss to rely exclusively on visitors’ remarks.
In other words, Zomato is not an authoritative guide in the everlasting pursuit of outstanding restaurants. Yes, it is a warehouse of valuable data, but it suffers from a set of inherent limitations. And you’re about to be in the know.
|Photo source: https://techcrunch.com/|
Just as with Facebook or Instagram, anyone with an email address can sign up for an account on Zomato. Which means that a restaurant owner can enlist his circle of friends and family to establish a Zomato username and leave a rave review in favor of his restaurant, thereby inflating the aggregate total. It happens ALL the time, and it’s so easy to spot!
Look first for the straight 5.0/5.0, and then check the total number of reviews and followers the account boasts. If both are ‘1,’ this marks the user’s first and only review and one sole follower, most likely the restaurant owner who of course also has an account. I like to call this the 5-1-1 trifecta.
Any time you’re scrolling through a collection of reviews and you stumble across the 5-1-1, take the review with a grain of salt. And quite honestly, it’s hard not to, when it’s entirely generic, lacking an ounce of detail about specific dishes, encounters with waiters, or evidence to corroborate a real visit.
|The 5-1-1 trifecta: A perfect 5.0 rating, 1 review, 1 follower. |
And that's the extent of the user's Zomato activity!
Many restaurants have caught on to the Zomato rating phenomenon, and even before welcoming paying customers, they stage a beautiful private performance, inviting the Top 25 Leaderboard of the local Zomato community. In other words, the most active and prolific users are hosted over a swoon-worthy culinary affair sure to elicit perfect 5.0s.
In one night, two dozen stellar reviews are logged onto the restaurant’s account page with Zomato, prompting a high aggregate rating before doors officially open. Anyone curious about the new restaurant would naturally sign onto Zomato and be overwhelmed with the effusive feedback, thereby spurring an onslaught of bookings.
These reviews are thus not reflective of a paying customer’s walk-in experience, and it’d be helpful (and honorable!) of the reviewer to mention as such.
How to identify solicited reviews? They usually occur in barrages, flooding the restaurant’s Zomato account over the span of 48 hours. And you’ll immediately recognize them by the Leaderboard’s usernames with title of “Connoisseur” in subscript.
In their quest to amass virtual points, rankings, and neighborhood conquests, Zomato reviewers often go hog-wild on reviewing and rating a large volume of restaurants in one sitting. A syndrome endemic to new users, such mass reviewing results in out-of-date feedback. These reviews lack lucid detail and photos, and their language is wishy-washy.
You can pinpoint outdated reviews by their reference to items no longer on the menu coupled with their generally shorter length.
|Outdated review: no reference to any concrete details; very generic, bland review|
It’s a well-known fact that those who take the time to articulate their feedback in a public or private forum are either motivated by an outstanding or awful experience. Restaurant visits that are not newsworthy—that is, they fall somewhere in the middle of the spectrum—don’t usually merit feedback.
Take that into consideration when navigating through Zomato reviews. These are the folks who, riled up by their positive or negative emotions, sought to praise or bash the eatery. Thus, reviews are not representative of an entire population sample. They are voluntary and self-sourced, and they should be treated as such.
|This user created a Zomato account to make known a poor restaurant experience|
|Another illustration of a user registering on Zomato to recount his |
extreme disappointment with an establishment
On a related note, the rating scheme on Zomato totally eludes me. One newly opened restaurant, for example, displays an aggregate rating of 2.8/5.0, which captures four votes, as the website claims. A solitary review exists, posting a rating of 5.0/5.0.
If that’s the case, the arithmetic mean is computed as (5.0 + x)/4 = 2.8, where x is the sum of the three remaining votes. Solve for x: x = 6.2. On Zomato, votes range from 1.0 to 5.0 in increments of 0.5, so x = 6.2 is mathematically impossible if you sum up votes. The decimal must end in either '0' or '5'.
So where did that 2.8 come from? Are scores somehow weighted based on the reviewer's heavyweight status on Zomato? If so, it's all the more reason composite scores cannot be taken at face value.