TLDR – The entire ratings system on seamless web is at best completely ineffective and at worst a set of lies intentionally ignored and even hidden by the company.
Let me start off with the email I sent Seamless Web 2 days ago. I had offered a UI improvement suggestion to them a few months earlier and they were very receptive to my comments so I decided to reply on the same thread. No reply yet –
“Hi John,
Hope you’re doing well. Thanks again for your accessibility last time around. I’m back to rant again unfortunately this time my complaint is somewhat more grave.
I usually do my ordering by selecting a cuisine type and then hitting the “Highest Rated” sorting option on the results page. Today I was in the mood for thai and i followed the same routine. As it turns out a restaurant called “Royal Thai Cuisine” was at the top of my list with a 5 star rating and 510 reviews as opposed to its next best competitor with a similar star level but only 230 reviews. Naturally I decided to go with the clear winner Royal Thai. Cut to me getting my food.
I could not take more than one bite of my food ( I am not prone to being dramatic. I am picky but NOT a drama queen ). The chicken was smelly and I literally was about to gag. Decided to throw the food and move on with my life. I did however decide to investigate how on earth such low quality food could enjoy the TOP position on your list of 22 restaurants in the area – 5 of which i have personally eaten at that I can guarantee have better food. One look at the ratings and review section and I was in shock. Had I looked at the recent reviews it would have been a clear red flag. Screenshot attached. It doesn’t take a genius to see that theres something seriously bad about the food at this restaurant.
Got me thinking what the problem in your rating system might be. I suppose that perhaps the quality of the food at this restaurant may have declined over time. It may have earned a high star rating early on and for whatever reason the numerous recent reviews that completely trash the place are not having any affect on it. I wonder if this means its time for seamless to tweak their filtering system so one we can choose how far back in time we want our rating statistics to be compiled over. ( I know this might be an issue with your data guys since ad hoc time ranges for large datasets can be computationally expensive ) however I know one thing for sure. Something needs to be done. Even no rating system at all is better than one that provides false positives.
Looking forward to your reply and really hope to see some change in this regard. Might be something as simple on the frontend as adding a “range slider” to review times in a users settings page.
Thanks a ton!”
I followed up the email with a link to the restaurants page on seamless and a screenshot of the ratings –
http://www.seamless.com/food-delivery/royal-thai-cuisine-new-york-city.9496.r?cm_sp=Consumer-Step-1-_-Order-History-_-Restaurant-Menu#reviewsbottom
It’s been two days so I decided to check back on the restaurant today to see if anything had changed. Guess what – the restuarant now has 696 ratings today ( what i had mistakenly called reviews in my last email ).
My best guess is someone is gaming the seamless web review system. A quick google for fake seamless web ratings and I’m starting to believe my own hypothesis.
Anyone with similar experiences?
armaanahluwalia.wordpress.com 