I pulled data on 1378 restaurants from Google Maps to rank them in order
mattsayar.com
mattsayar.com
I would say the best places typically score low fours and have at least some one-star reviews written by seemingly deranged people.
One might think that caring about not having bad reviews corelates with caring about the quality of ones services, but I haven' t found this to be the case.
I've noticed the occasional incredibly bad business that gets bad reviews removed, but always wondered how they did it and how that works? A guess: they make some fake google accounts and report the review from multiple accounts?
Unsure whether they're actually capable of doing it or not. But they exist.
It's heavily dependent on the area in my experience. In some places 4.3 restaurants are perfectly fine, in other it's at least 4.5. Which is why when I'm travelling I don't simply visit the first high-scoring restaurant but look closer at several ones to get a general idea of the scoring habits there.
If those are complains like "waiter was not nice", "they took 30 minutes to bring my food", I'll assume there's nothing to worry about.
My feeling is that bad reviews are more likely to be more trustworthy, because they are less likely to be faked (unless its the competition).
Look at how many reviews will just be five-star ones, but no text. Or, one or two words quips, like, "great food!"
Negative reviews might also be emotionally driven, some by one-time events, like a to-go order taking too long because of a large office order that preceded it.
When I was apartment shopping, I saw so many "sketchy" reviews on Google and reporting them does nothing. Many would be from people who only toured the place and NEVER actually lived there, or from people who just moved in. Both are useless, but does Google care? No! Hence, my disdain for Google reviews.
Amazon Fake Reviews also stopped being reliable because people figured out how to game it.
What do you use as alternative?
What you can do is go through the text reviews and try to read between the lines and figure out if the place/item is good for you.
How do you make money without skewing the results?
And can LLMs be subtle enough?
Last time I picked a restaurant via reading between the lines, it was a negative review that made me decide to go there. The person leaving it was either used to fast food like service or was in a real hurry, but between the lines the food was great, they just took a while to serve it. I was in no particular hurry so I was happy with the place.
Can a LLM figure that out? And what about the day when I am in a hurry and I am looking for fast food like service?
https://www.top-rated.online/ re-sorts the reviews (on individual place page) based on reactions and number of reviews user made. It makes it easier to see the full picture and avoid fake reviews.
Firstly, nice site - always love new tools to discover restaurants, thanks for posting, I’ve shared your blog post with friends, it was a brilliant read.
I have some recent experience working with restaurant reviews, I found that using only Google reviews can be unreliable, as some places that have top reviews may not be generally accepted as the ‘best’ restaurants.
We currently use a combination of Google reviews + Trip Advisor + Reviews from the booking platforms and we have web crawlers to check if the restaurant is featured on reputable restaurant guides or review sites.
We aggregate all of this review data and compute a “score”, so when users search for available tables in a city we can show available tables at the highest scoring restaurants first.
We apply Wilson score confidence intervals, to trust restaurant scores that have more reviews.
We are also applying an exponential decay when users list nearby restaurants, as you might be willing to travel a little further to go to a higher scoring restaurant.
Working with review data is fascinating.. we’re going to be launching an AI summary of recent reviews and our computed score in the coming weeks to help our users understand our ratings.
Our app went live on the App Store only a few days ago and we expect it to be live on Google play later this week.. so it’s an extremely busy time!
If you’re interested in what we’re doing please reach out, it would be great to connect, I really enjoyed your article!
The Cuban places were upscale, fancy, had amazing food and drinks. They became the de facto 'take a visitor out' spot. No qualms there with the ranking, few places even seemed to compete.
The disconnect(ie funny?) is that Americans think of Cuba as a failed, second/third world country as we're taught. So that would seemingly lead to the people/restaurants owners being cheap and scrappy. But here they are, showing everyone else up in multiple cities. I respect that a lot.
3.2.3 Restrictions Against Misusing the Services.
(a) No Scraping. Customer will not export, extract, or otherwise scrape Google Maps Content for use outside the Services. For example, Customer will not: (i) pre-fetch, index, store, reshare, or rehost Google Maps Content outside the services; (ii) bulk download Google Maps tiles, Street View images, geocodes, directions, distance matrix results, roads information, places information, elevation values, and time zone details; (iii) copy and save business names, addresses, or user reviews; or (iv) use Google Maps Content with text-to-speech services.
Foursquare released its database of places, maybe that would be more interesting to OP (as well as the data from OSM).
Foursquare's 104M Points of Interest https://news.ycombinator.com/item?id=42219578
Foursquare Open Source Places: A new foundational dataset https://news.ycombinator.com/item?id=42191781
Demo of the dataset: https://wipfli.github.io/foursquare-os-places-pmtiles/#map=1...
I stopped maintaining the project 1 year ago so the list are getting stale now but it was fun while it lasted. Glad to see someone else look into this!
The crappiness of the ranking and filtering options in Maps search is completely inexcusable. Ranking is Google's core business!
Select your city and sort by "Most Reviewed." I use this site all the time to decide where to go when visiting a new city -- it’s incredibly useful.
https://www.evanmiller.org/ranking-items-with-star-ratings.h...
All Evan Miller's posts on user ratings are excellent - https://www.evanmiller.org/ and search for "Mathematics of user ratings".
https://wanderlog.com/list/geoCategory/1/where-to-eat-best-r...
Looks like Boulder County has one too https://opendata-bouldercounty.hub.arcgis.com/documents/c9d2...
I was in Spain for a couple of weeks and every 4.5 or higher rated place on Google maps was a disappointment and had handsome nice waiters
It's amazing how many people are happy with expensive poor food with nice service (I do understand service is part of eating out of course)
I'm pretty sure Google Terms prohibits using data from its Maps API otherwise than in connection with displaying/using a Google Maps service, but maybe they won't go after this, because it's not much data.