Good for her. Yelp's business model seems to work best if they can pick winners, which is unfortunate.
If I search for "dinner" on Yelp for "current location" and look at the map, it leaves out an astonishing number of places that I can see within a mile if I drive down the street.
And if your business model is centered around hiding all but the winners - who might pay for placement + reviews both - then you have to try to keep people from talking about it, and the moral interests just get worse and worse.
Google Maps isn't any better; OpenTable seems the "least bad" of options I've looked at recently, at least for fancier reservation-having places, in terms of actually showing my options.
(It's a broader internet problem in general - "trending" algorithmic reinforcement stuff will push everyone towards the ONE TRUE BEST HOTTEST MOST POPULAR bbq/ramen/whatver place, when probably there's half a dozen other options within 10% of taste and quality, that some people very well could legitimately like better if they heard about them too.)
Google Maps is terrible for hiding results. I'll use the filter to show restaurants with a 4.5+ star rating, and it hides dozens of places 4.5+ stars while instead showing results 4.0 - 4.4 stars and even some that have no reviews at all. I find the arrogance insulting. You want to see 4.5+ star Japanese food restaurants? Well, I don't think that's best for you, and I'm going to replace some of those results with 4.1 star Caribbean food restaurants that you'll like instead.
I despise Google Maps for this. I can know that a restaurant exists, zoom in on its block, search 'restaurants' -- nothing. It'll zoom out the map and show other stuff. Worse, I can max my zoom and hunt for it on the map and it will outright show a blank space. Only searching for the restaurant's actual name will bring it up.
But there's not really any alternative. Apple Maps can't show me things because it's missing the entries entirely. Yelp became useless years ago. I feel like we're going back in time because the best way to find local restaurants has become walking / driving around the neighborhood to see what's there, or getting word of mouth recommendations.
Internet search in general has severely declined in 2023. I was trying to figure out the syntax for putting a URL in slack the other day and couldn't find a proper article anywhere in the top 10 results.
I think they are trying to integrate their bard UI into search and it is killing the results.
At this point, I use google to search reddit in order to find organic results. Unfortunately, that will probably be taken over by spammers soon enough.
YouTube is also horrible these days. They made it so for any event or happening, only "official" sources show up so they can add "context" to the narrative. It's dystopian, but possibly worse: annoying. I can't find out what I nees to, when I need to. Thanks Google.
Or they outright silence smaller news organizations that broadcast popular, although controversial, figures who fall under a certain political affiliation.
I'm grateful for places like Rumble and Twitter, where real news can be found. :)
> I feel like we're going back in time because the best way to find local restaurants has become walking / driving around the neighborhood to see what's there, or getting word of mouth recommendations.
Let's face it, that never stopped being the best way. There have always been problems with internet reviews, if only the ones inherent in the 1-5 star mechanism. But, it is very inconvenient to ask around for recommendations when you're (for example) hungry and passing through an unknown town or something. So I get why we need a source that isn't completely unreliable.
Perhaps what's needed is a website that doesn't do reviews, but only lists businesses that are open to the public and categorizes them by function (pharmacy, grocery, daycare, etc). Possibly with direct links to review platforms and official websites and so on, but without actually inserting any content (reviews, ad copy) into the website itself; this can be left to other sites.
I've noticed this in the last 6 months specifically. If I search for grocery stores it only shows me some chains and not others now. If I want to see where Safeway stores are at I have to search Safeway; they won't show up under grocery stores anymore.
I have noticed this too. I wonder if those businesses haven't paid an SEO expert to set up their business properly or if they haven't paid google for their "premium" map features
Organic Maps is very good at showing where 'restaurants' are. It's just not very good at showing their hours or including links to their web site or menu.
Google maps is also terrible at showing _only_ relevant info. I searched for “coffee” in a small area earlier today (small because I wanted to walk); it zoomed out to a much larger area and showed me coffee shops along with taquerias, oil change shops, banks, etc. both in the results pane and scattered all over the map. It was such a mess as to be unusable.
Thank you, this is a huge problem with maps search results. It's showing too many results and the ranking is rubbish. Searching for coffee shows tons of low quality results including ones for Target and McDonald's.
You're being flagged but you're absolutely correct. Maps search is awful. Yes, most McDonalds might have more than 4 stars, but that's not what I want when I search 'cafe'! I avoid most chain restaurants in general which makes it much worse. I wish there was at least an option to block restaurants from showing up in your results.
on the same topic - sometimes it's so frustrating when I put some shop name in Google Maps while I'm in Paris and looking on the map of my neighborhood, for example - and whoosh - it zooms so far out to show all the places which "suit" the search in all the Europe. Why would it do that?
It's so frustrating to find the perfect restaurant by chance and find it on the map directly when it didn't show up in a very targeted search of the same map.
There's also a lot of brigading on Maps. You'll see 5 1-star reviews in a row, and realize they are either duplicate accounts or friends who are all reviewing the same experience.
Weird thing is I began comparing things like Uber Eats reviews and Google Maps reviews for local Asian restaurants, and I see things in Google Maps are reviewed substantially lower.
For example some of my favorite Asian restaurants in the city are rated 3.5ish while Uber Eats is rated like 4.7. It leads me to believe a lot of people are rating Google Maps restaurants also based on service. One in particular
This place is awesome, and personally have experienced bad service but don't really care, so I do understand the Google Maps rating, but do find the whole rating system not helpful for me as I'm interested in good food and willing to accept bad service.
One theory I have is that you're getting different rating for different experience. Some "Asian" food keeps well, so it could stand out from the competition. They don't have to deal with you being in the restaurant, complaining about the smell, the service, they hit on the waitress and she wasn't very friendly responding, ect.
You're correct. This is a problem with ratings. You really have to read into them to see what they're complaining about or praising. This is true even for individual reviews, written solo. Many are unhelpful, or stuff that doesn't concern you.
One of the best examples I can remember is a venue that had 2-3 stars. However, all the negative reviews were like: "They have massive parties going hard late into the night... but my kids go to bed at 9:30, so 1 star".
> Well, I don't think that's best for you, and I'm going to replace some of those results with 4.1 star Caribbean food restaurants that you'll like instead.
If only. Its more like "these are the businesses that have paid me. They don't match what you want, but tough cookies."
Same with the much vaunted Amazon and Netflix recommendation engines. They stopped recommending stuff that you would like a long time ago, around the time everyone was trying to copy their recommendation engines, and started recommending stuff that makes the company more money.
I hesitate to suggest that this may indicate there aren't that many 4.5+ star Japanese food restaurants. Even great resaurants in my area struggle to break the 4.5 barrier.
Changing the zoom level will typically show these missing results. For example, I just checked my neighborhood. There are 5 restaurants within a few hundred feet of each other rated 4.5+ stars. With my filters set to 4.5+ stars, when I zoom out a bit, 4 of the 5 restaurants disappear, and I instead see a handful of 4.1 and 4.3 star restaurants. One of the 4.3 star restaurants is literally next door to a 5 star and 4.5 star restaurant, which are hidden at that zoom level.
I thought maybe it's some kind of weighted rating, because it's showing me the 4.3 star restaurant with 1,100 reviews, instead of the 4.5 star restaurant with 600 reviews. So, perhaps Google thinks the additional number of reviews make up for the lower rating, and they rank it higher, even though I'm specifically asking to not see those results. However, it also shows a 4.4 star restaurant with 40 reviews, while hiding a 4.5 star restaurant with 90 reviews.
This is par for the course when it comes to modern recommendations though. Do you want to see Netflix movies released in the past 3 years? Or what about a simple list of top rated movies? Too bad, you can watch what we tell you to watch. It's a complete lack of respect for the user.
It could also be recent ratings; I think it is reasonable to weight old reviews lower than new ones given restaurants can change a lot over the months and years.
Perhaps. But then why not just say “no results found”? Because that might give the impression to the “average” user that Google is broken and isn’t solving their problem. To the more sophisticated user, Google is just meddling where they shouldn’t.
Google is proving the central limit theorem by catering to the average.
Google long ago decided that "No results found" is never an acceptable response. Same with regular Google search. It will just choose arbitrary things to show you instead.
It's one of the things that has reduced the usefulness of these tools.
At least here where I live, Foodora and similar have ruined what was left of the Google ratings for any place offering take-away, because there's ton of 1 star reviews which concerns the transport.
At least viewing the most recent comments and reading the comment that goes with it helps figuring out if it might be a decent place or not... at least sometimes.
It gets worse. People have been 'following' the ratings when they rate foods.
A popular local restaurant came under new management and their food is now atrocious. We actually threw their food in the garbage after getting takeout there a month ago.
Meanwhile, tourists are still leaving thousands of rave reviews on their google maps page. Noticeably, you will see one or two bad reviews every 50 reviews or so, where a person comments about their confusion about the restaurants score.
Yep, a local pub used to give large portion sizes on in person or takeout orders. Then one day we ordered and it was like half the size as normal but of course the same price. Thought it was maybe just a fluke so we gave them a second chance. Unfortunately the portions will still the same smaller size. I proceeded to update my review to indicate the change in portion sizes with the same price. However I noticed many older reviews still indicating large portion sizes as well as even new reviews saying it. Because many new people wouldn't even have the comparison to how it used to be.
So you're getting both new reviewers (or travelers) who don't know it used to be better who still call it good, mixed with old reviews calling it good.
It's not even just a management change that causes this. This is when you almost wish Google maps would put your location history to use. It sees you visit the place frequently and thus if you stop visiting and leave a negative review it should weigh that review higher and give it more visibility. An old stale review or a review from someone who visited once I don't think should be as impactful or useful on the experience.
Overall I find discovering new places to be frustrating as reviews often seem useless. Usually I find the best method of discovering new places is still via word-of-mouth in person.
> This is when you almost wish Google maps would put your location history to use. It sees you visit the place frequently and thus if you stop visiting and leave a negative review it should weigh that review higher and give it more visibility.
But how would that make Google more money, this quarter?
I could imagine an argument that, over a long enough time period, (re)gaining user trust would lead to an increase in ads revenue, but at best that long-term increase goes into the promo packet of the T9 leading "Geo/Commerce" or whatever the product area that contains Maps is these days.
DoorDash shows popular items purchased from restaurants along with ratings for each item, plus some form of recency for purchases like "Ordered recently by 20+ others", even down to the options / toppings selected for an item.
“The problem is, or rather one of the problems, for there are many, a sizeable proportion of which are continually clogging up the civil, commercial, and criminal courts in all areas of the Galaxy, and especially, where possible, the more corrupt ones, this.
The previous sentence makes sense. That is not the problem.
This is:
Change.
Read it through again and you'll get it.”
-- Douglas Adams, So Long, and Thanks for All the Fish
Any information repository eventually comes to face the problem that the phenomena it is describing is no longer the same, for intrinsic or extrinsic reasons.
> Yelp's business model seems to work best if they can pick winners,
People must have read a different article than I did. This woman is finding tons of third-parties that are either posting fake positive reviews for money, or extorting businesses by writing bad reviews and offering to reverse them for money. She's looking for people who are trying to scam Yelp, or trying to scam businesses through Yelp.
But HN seems stuck in a narrative (for a decade, not new) that Yelp is some sort of organized crime syndicate, and refuses to budge.
"Picking winners" here doesn't mean Yelp is doing the faking themselves.
I just mean that Yelp has chosen to filter and limit in addition to what I'm actually searching. It might be a conversion thing - if I have one really good match it's easier to decide than if they actually show me all 10.
But it means it's really really important to the business now to be that 1 that doesn't get filtered.
And THAT really increases the market for all these fake-review things.
The article makes it pretty clear that the villian is Yelp (and Google Reviews, and Trustpilot etc). It's no different than any other social media moral hazard: Yelp makes money in spite of their being fraudulent reviews, and won't make the investment in cleaning it up.
> Dean said [Yelp's] notices validate her work, but also perfectly exemplify what she calls Yelp’s “whack-a-mole” approach ... she showed SFGATE several sketchy posts that she hasn’t mentioned in her videos — and which are still up and active.
> “I find it annoying, like, ‘You need me?’” she said. “... They’re a billion-dollar tech company that’s got teams of engineers and a trust and safety team.”
> ... Yelp said its automated recommendation software checked the 21 million reviews submitted last year for “quality, reliability and user activity” and had filed 75% into the “recommended” category, which are the reviews that figure most prominently on businesses’ pages and affect their star ratings. Of the remaining 25%, just 4% were removed by Yelp’s own moderators, 2% were removed by the reviewers themselves, and 19% were categorized as “not recommended reviews” — which don’t affect companies’ star ratings but are still accessible on businesses’ Yelp pages.
> In Dean’s mind, it’s not enough. She feels Yelp should better use account data to proactively identify problematic users and ban them altogether, and speculates that Yelp might be avoiding a crackdown on fake reviews because the positive boosts to businesses’ ratings might increase Yelp’s value to those businesses.
Given that this kind of problem is rampant across social media (where the platform is incentivized to promote horrible things, as we've seen with Meta, Insta, etc etc), it probably should be criminal if we ever want to erase this scourge. As the article shows, _businesses can't opt out from Yelp_ -- which is a pretty horrible state of affairs.
> It's a broader internet problem in general - "trending" algorithmic reinforcement stuff will push everyone towards the ONE TRUE BEST HOTTEST MOST POPULAR bbq/ramen/whatver place, when probably there's half a dozen other options within 10% of taste and quality, that some people very well could legitimately like better if they heard about them too.
Instagram doesn't help -- gotta get a shot of me eating noodles at the hot new place! There's a new ramen shop in downtown San Mateo that has a 30-person line outside at all times... I can't imagine it's that much better than the half-dozen other ramen places within 3 blocks, but by god it's new, and anyway would people be lining up like this if it wasn't good? I better get in line too...
Back in my bar-tending days we always told the door staff (bouncers) to slow down when checking IDs early in the evening, just to get a line going. It works surprisingly well for drawing in more customers.
Totally tangential but follow-ups on that psychology:
* some people would walk in and see the still mostly-empty bar and get fed up that they waited in line for "nothing". This was surprisingly few people though.
* some people would ask and buy just about any explanation - "oh that guy is new and still learning", "we had someone get in on a fake ID and got a fine the other day so the guys are just being careful" or whatever and the people would be glad they got in early to get a good spot.
* sometimes we'd slow down the bar staff a bit too so that there was a line to buy drinks and people were happier that they had to wait in line again - confirming their bias that they must have gotten in at the start of the rush.
* some people were wise to the scam and everyone who called us on it thought it was great. They often offered to have their buddies come through the line a few times in exchange for a free drink.
* for some reason when you let your buddies, regulars and off-shift staff cut the line and get in quick, the rest of the folks seemed more willing to wait in the line.
Utterly bizarre, yes. I think the psychological term for this is "social proof", and it applies to all sorts of weird things. For instance -- a person who is getting lots of flirty attention in public will tend to appear more attractive to others. Not because they are, but because seeing them get attention is social proof that they're worthy of attention.
Once I became an old fart, I started seeing these lines as a sign to move on. The question in my mind (at any age) was always "is the place worth waiting in line for?" But these days, I already know from years of experience that the answer is very likely "no".
Skipping the line has improved my nights out, though, because in looking for an alternative, I frequently find hidden gems and can enjoy them before the rest of the city discovers them.
This has been a process for me. I still occasionally succumb to the silliness, but for the most part when I see a line, my reaction is, "meh, I'm hungry/thirsty now; let's go somewhere else". One thing that's telling is that some of the places I already know I really enjoy rarely have lines or are completely busy.
And absolutely agree that you often find hidden gems that way. A lack of a line will often just mean the business is too inexperienced or honest to use some of the psychological manipulation practices the GP talks about.
I recently went to a dessert shop in Palo Alto that seemed to be rate-limiting their orders so they could get a crowd formed. Once there were 2-3 people waiting, the line would invariably blossom to 5-10, and then stay high for a while. I can't be sure it was on purpose, but the thought definitely crossed my mind...
I know which place you're talking about, and their stuff is actually sort of unique to me at least, in the sense that there's no broth in any of the ramen, but more of an oily sauce that you mix with the noodles. It's actually really amazing, and worth at least some wait to me. (and a Japanese opening a bay area branch tends to form lines).
This is why Craigslist does great. They take a small fixed fee on volume and don't favor listings.
A good restaurant directory - some exist for restaurant niches, but none in general - could clean up with non-obtrusive ads if the focused on completeness and accuracy with a fairly minor feature set. It would just require having enough capital and time to establish social trust.
Curating reviews is just broken, just like SEO has turned out to be. It is unavoidably gameable. Yelp seems like a zombie site. They are free-riding on their prior reputation and gradually spending it away to acquire revenue. When the reputation is gone, the revenue will be, also.
Founded pre-internet. Curated reviews behind a paywall :. clear incentives for quality. A quick glance over wikipedia gives the gist. In my lifetime Zagat-rated was a mark of honor particularly for mid to high-end restaurants.
I'm always amazed when I search for any generic term on Google Maps at how many places I know about don't show up in search results. What a shame. They clearly ran out of new ideas years ago.
First time I used OpenTable I went to a highly rated and expensive place, had horrible food, and then my wife spent most of mothers day in bed with food poisoning. 0/10.
> Dean began documenting her findings, starting with serial offenders from the Facebook groups. In a list of 17 people who reviewed SavantCare on Yelp, she found what she called “a suspicious pattern”: 10 of them had reviewed the same dentist’s office in Los Angeles, nine had raved about the same California wedding DJ company and five claimed to have visited the same Mountain View tattoo parlor. Five of the accounts, she noticed, had even reviewed the same moving company in Florida.
SMH... I mean, how difficult is it for Yelp to catch this?!? Hire a high school student with interest in data analysis, and you'll catch these idiots quickly.
Which turned out to be a terrible idea, because now with the importance of data integrity in training AI all these BS reviews everywhere mean Yelp, Amazon, etc are left with effectively worthless data that otherwise could have created very neat tools and offered a secondary revenue stream to just placement/ads.
"I mean, how difficult is it for Yelp to catch this?!?"
The mistake you're making is assuming Yelp:
* Wants to "fix" this problem.
* Isn't actually complicit in this problem as a means to incentivize small business owners to sign up and pay for Yelp to help "manage their business profile".
And the second mistake is (if they did care) thinking this wouldn't just result in an arms race. IMHO we want easy to spot fake reviews. I want flash to come back. It was so easy to block annoyances in that day. Now we have HTML5 popups.
I assume you're talking about it being easy for the consumer. There is no way I would know that the same college kid copy pasted a pic from a Flordia Zilliow listing to out on a home renovation review but it's trivial to find for Yelp. So I understand the sentiment, but at least to my untrained eye, I just glance at reviews and take them as real as long as a dog walker is not mentioned at a sushi restaurant, which means they are too hard to spot for me.
They probably make more money if reviews for a restaurant are in question. If someone spends 20 minutes clicking around on a list of restaurants to read the reviews and see if some seems genuine or fake... yelp makes lots of money.
If yelp reviews were 100% trusted, everyone would just click once, and then go to the highest rated restaurant.
They should also not be allowed to make secret deals with companies to artificially boost said companies or to downrank non-paying competitors.
The problem is there's no accountability.
Sure, we can sue them, but for what damages? How do you account for the value of lost business opportunities caused by Yelp selling your honestly earned ranking spot on their website to your competitor?
The only way to prove they've done something wrong is to commit corporate espionage and get the proof yourself, which would then most likely invalidate the ability to use that evidence for revenge.
Yelp and friends DO catch tons of this, but the people doing it keep working and working until they get past the filter.
Whereas if 1024core or bombcar posts a review, and it gets blocked/shitcanned/shadowbanned/ignored/never posted, we give up and move on with our lives, we have better things to do.
Most of the "Yelp is extortion" line comes from businesses that have had good reviews hidden (that they probably gave people discounts for, or paid for.) Every good hidden review is seen as an outrage.
It'd be less controversial to kill bad reviews like this (because bad reviews often can be extortion or attacks on competitors), but the users would notice and also bitch. [edit: even worse, businesses that got bad reviews deleted by talking to Yelp might be characterized as "in bed with" Yelp.] A Yelp-like business is no-win imo. Voting isn't much of a signal if you let anybody who happens to walk by vote.
IIRC Dan Luu had a post about this: sites do catch a lot of this but there's just so much out there that they can't catch it all, not without accidentally flagging genuine reviews and getting everyone mad at them.
They do flag tons of reviews (I think I’ve successfully had one “not five stars” review go across on ANY site, period, out of multiple tries. I don’t even bother anymore.)
It's pretty good stuff. She builds a devastating case very methodically, showing how the identical glowing review text is posted to yelp, to FB, and to Google, by different users -- and then traces other reviews posted by those same user ids showing how they are plagiarizing Tripadvisor posts.
I am surprised a yelp/google alternative has not been created with a feature to establish provenance of a review. Perhaps something like a receipt which contains a unique link and that person can provide a review through the link.
The review will be posted as "verified transaction" on the site and overall score of the restaurant is determined from there. "Anonymous" reviews can still be posted but won't be factored into the overall score and will be deprioritized by the system/algorithm.
Of course this does not necessarily stop businesses from pumping out fake transaction tickets and offering users to write fake reviews. But at least a paper trail would be established; and if a historically 1-2 star business with verified transactions and 1-2 reviews per day suddenly sees a spike to 4-5 stars with 10+ reviews per day over the same period. Then it can be flagged as suspicious and team can review the paper trail.
I actually worked for a startup that did this, albeit it wasn't public-facing, rather we'd incentivize people to leave reviews in exchange for a coupon code or whatever
Turns out a lot of restaurant owners would prefer to straight up own the reviews/influence it themselves rather than get legitimate, anonymized ones. There's just not that much of a reason to genuinely improve as a service when there's so much cutthroat competition
Not sure, I left quite a while ago, haven't really kept up with them. I doubt it though, the CEO didn't care too much about infinite growth and just wanted to build a helpful app for the handful of people who would appreciate a service like that
Retailers do this on their own stores via "mystery shoppers"
The retailer pays a firm to go in shop at the stores and they write a detailed report of customer service, how clean the store was etc.
You need restaurant's and the reviewer's cooperation, and it seems your proposition would to have the restaurant foot the bill. Which means, they pay to you give more control over their reviews and ultimately their business to random customers.
The user spend more time and effort to leave a reviews, and while they might see some side benefit in improving your service, they now have to trust you won't screw them and sell them to the restaurant.
Do you see a third party willing to pay enough to keep you alive, outside of spam and ads ?
> A Yelp post from December 2017 was the first crucial clue. “Melissa R.” wrote, “I was offered money to leave a positive review for Savant Care Clinic. Must be a shady company.” The user attached Craigslist screenshots of the apparent offer, which Dean took her own screenshot of just before Yelp removed the post for violating its terms of service.
Why would Yelp forbid people from talking about this? While it is certainly possible to create fake email screenshots, this seems like a relatively small risk compared to the much more common risk that's being called out here.
I guess they just don't want people to know how common this type of occurrence is, since it would undermine all trust in their platform.
The risk (mostly) isn't fake email screenshots, the risk is that I create postings for (or even commit to paying for fake reviews for) my competitor, and tarnish their reputation. It's hard to establish provenance of such an arrangement.
I think we're thinking the same thing — that someone would (on their own or for money) write a review claiming that an innocent business had tried to pay for a good review, or take down a bad review. This could happen, but it's just as easy to pay someone to write a review saying the service/food/etc. was lousy. Why spend more time faking an email screenshot?
But from Yelp's perspective, a paid-for review claiming bad food is just fine, because it only tarnishes the restaurant's reputation. On the other hand, a claim of fraudulent review manipulation hits both the business and Yelp's reputations.
I think they tend to interpret anything that makes Yelp look bad as a violation. The owner of a wine store I bought thousands of dollars of wine from was arrested for running a $45 million Ponzi scheme (and eventually went to prison for years). I immediately put up a review, and Yelp took it down because the arrest wasn't my personal experience. Never mind that I had to get refunded by my credit card company. I was doing a public service, as a customer, and Yelp wouldn't let me. I took down all my reviews and deleted my account.
I'm guessing the ToS violation was that the review wasn't about an actual visit to the business, but was just a report of a Cragslist ad the user saw about buying reviews.
I agree that this sort of information is useful for potential patrons to have, but can understand that Yelp wants their reviews to be about actual visits.
> Meanwhile, Dean took SavantCare to small claims court over the letter and calls she later received, and won a $275 judgment in March 2018 for “negligent infliction of emotional distress.”
NIED claims are nearly impossible to win. In law school we were taught to point them out on issue-spotting questions, but that they were basically unwinnable.
This is an interesting thread, so I thought I'd give some more detail. A few months earlier, my husband and I visited small claims court and had the opportunity to ask a judge if I could bring an emotional distress case there. His response? "Sure you can." We poured through the CACIs and prepared our case for weeks, knowing how difficult it would be to prove. Frankly, I thought we had enough to prove intentional emotional distress, but I was satisfied with getting an award for negligent. Yes, Vikas Kedia and the two head doctors for Savant Care showed up. At the time, I didn't even know that the letter from the lawyer had been forged! Savant Care knew it, though, and hid that fraud in small claims court. Regardless, I still had enough evidence that a psychiatric practice had inflicted emotional distress that I won the case. As the article mentioned, it was about principle, not money. Savant Care eventually paid, but only after we sent them a letter threatening to drag them back to court. Of course, this case only concerned Savant Care's actions toward me, not their fake reviews. Kay
That's either not the case (this was in California), or is for something "worse" than negligent infliction. As a sibling comment pointed out it's possible that this was a summary judgment so perhaps it isn't technically emotional distress.
Not egregious, certainly for the couple hundred dollars recovered. I was mostly pointing it out (1) because any lawyers here would find it interesting that someone, somewhere, once won on an NIED claim, and (2) so that non-lawyers wouldn't get the sense that NIED claims are winnable, in general. They are often tacked on to other, more legit claims, but they are very rarely won.
Egregious in the sense that it was so over the top obvious there was no other outcome but to award the win vs some minor squabble between two parties. Not that it was so egregious mountains of cash were expected.
Not sure of the jurisdiction issues where the judgement was issued but the last time I looked into it you can't meaningfully collect a judgement from small claims court in my area.
Seeing the 'Moving company from Florida' triggered me. I was scammed by a moving company from Florida in 2020, and I did a deep dive to look into them and I found huge networks of moving companies, scammers, & reviewers, all in cahoots.
I actually managed to get one company's fake reviews removed from google maps, bringing them from a 4.8+ to a 1.x overnight. Within a few months they were back to > 4, but the owners had sold the company and started a new one.
It's insane how deep this goes. At best it's done by companies that want to cheat the system by getting an unfairly good review, but at its worst it's used to trick users into a false sense of confidence so that they can be scammed.
Moving companies, locksmiths, plumbers, tow trucks. These businesses all have certain things in common:
1) They work in business sectors which you usually only need rarely, so there's no need to build customer loyalty.
2) But when you do need help from one of these businesses, you need it now. There's little room for a customer to negotiate or shop around.
3) As a result, online advertisements and reviews for these businesses are an utter cesspool. There are very few honest reviews, and it's common for "local" business listings to actually connect you with brokers.
Same with tow trucks. Trying to find a tow truck on your own is a HUGE mess. Finding a tow company in the area invariably leads to some sort of "virtual tow truck call center" that takes a bunch of money and subcontracts out to actual tow truck drivers.
My solution was to pay for the roadside assistance via my car insurance and call them to sort things out.
Now emergency plumbers - same mess, but I haven't figured out a solution for that.
Same with locksmiths. Where there's clients in need of emergency services, there are always grifters willing to take advantage of desperation.
(Use AAA for tows and vehicle lockouts, never a locksmith you find on your own.)
Unlike roadside assistance (which works), the one thing I can tell you not to do for emergency plumbers is trust a home warranty service like American Home Shield. These sorts of "services" pay contractors subpar and flat rates, which deprioritizes your claims and incentivizes corner-cutting. They will not deliver miracles like finding you a plumber on Thanksgiving.
> Before Fake Review Watch and all her research, Dean used review sites often, she said. But now she tells everyone she knows to stick with good old word-of-mouth — especially with recent additional hits to the online system, like the surge of reviews spurred by COVID-19 policies and new reports of AI chatbot-written Amazon reviews.
Kind of sick of reading this advice. If I knew someone I trusted to recommend me a contractor why would I be on Yelp looking in the first place? Hello.
This is where NextDoor comes in. I guarantee your entire town is likely already on there talking about contractors they've used. The app is slow and sluggish but it's legitimately where most homeowners and "regular townsie folk" are these days.
I am on some Facebook neighborhood groups which I think are similar enough to Nextdoor, but there are plenty of scammers about and also, it’s hard to trust people’s judgment given how many idiotic posts you see floating around.
It sounds to me like Yelp is upset that other businesses are trying to do the same thing they do. My friend was GM of a new local restaurant here, and at first they were getting lots of 4 and 5 star reviews and almost none below that. But around the time they were contacted by a Yelp sales rep to start paying for promotion and advertising, the 4 and 5 star review stopped coming. My friend watched multiple people post reviews that never appeared. Then negative reviews started appearing from people that he knew never came in because he was working the host station all day. All this activity was reversed once they started paying for Yelp services. It's the kind of behavior that's impossible to prove for anyone outside of Yelp, and they of course claim they don't do it.
Same story from a restaurant owning friend except the sales rep called him a second time offering to remove negative reviews as a service. Simple extortion like the mafia.
I've heard the same stories from several local restaurants. With one added detail, which was consistent between the 3 restaurants: The sales pitch was always $300/mo. to ensure they have only high ratings. When they don't pay the fee, the negative reviews start showing up, and the positive reviews start disappearing.
It's such a consistent story it beggars belief that there's nothing there.
It's normal for a new local restaurant to get a bunch of good reviews because locals are trying it for the first time and want to help. Then it's normal for reviews to drop off because every local who would leave a review, has. Then it's normal to start attracting negative reviews from people who have never visited because local competing restaurants have realized where the downtick in their own business has come from, and so start trashing the competition. And then the negative reviews stop because the competing restaurants were satisfied they'd left enough negative reviews.
These stories have been going around about Yelp for over a decade at this point. And Yelp has churned through so many salespeople... you'd think there would have been tons of credible whistleblowers by now if this were really how Yelp operated.
Apparently Yelp will remove positive reviews that have been posted before, and save up negative reviews for publication after the advertising pitch; to be unpublished if advertising is paid for.
They may not actually be writing the reviews, simply choosing when to publish or unpublish them, and in what quantities, and which to count for the overall rating.
It doesn't matter who's casting the reviews, but who's counting them.
If that’s the case shouldn’t there be at least one screenshot of this behavior happening? I’ve heard dozens of anecdotes over the years. You think that at least one restaurant owner would screenshot their rating and the number of reviews as soon as Yelp reaches out to ‘suggest’ advertising.
The fact that this seems to be commonly-accepted knowledge without a single piece of evidence is strange to me.
In fairness running a restaurant is nearly a round the clock job. Running delivery and the online reservations is more. Managing reviews is even more.
I agree that its hard to believe one restaurant hasn't tracked it yet but I think its more a confluence
- are you good enough/ no dip in performance that you can claim this correlation or are you confident enough to go out saying that knowing you have to defend against ppl saying "well maybe you just suck now!"
- did you have time to screenshot it when it was happening aka the day Yelp called and your dishwasher was out sick and your supplier of eggs upped their price
- how long does the business even run. lots of restaurants close down and Yelp is not on a top ten list of concerns
- if you're established then your focus is growing business not dealing with the mafia. maybe you pay them, maybe you double down on other channels like uber eats or a new location or whatever.
Sure, that's all totally fair, but was there seriously not even one -- out of the hundreds of thousands (millions?) of businesses -- business owner who managed to find the time to do this?
> You'd think there would be some whistleblowers after all of these years...
I wouldn't think so. In a society so rife with obvious scams everywhere you look, at every level in public and private sector orgs, it would be patently idiotic to trash your own career on account of a dishonest restaurant review operation.
I wish we lived in a world where such bravery were meaningfully valued, but we don't. For any might-be whistleblowers, this reality has been made painfully clear in the time since the Snowden leaks nearly a decade ago, during which nothing has changed.
The thing about whistleblowers is that nobody actually likes them. "Whistleblower" is to white-collar crime as "snitching" is to anything street-level.
We "appreciate" them when they call foul on others, but you appreciate them a hell of a lot less when you have such moral crusaders in your own ranks. They aren't loyal to anybody but their own sense of self-righteousness and can't be trusted to keep their mouth shut if you piss them off after they discover dirt on you. They're only ever a liability.
On the flip side, barring being vindictive it's not worth it to blow the whistle on anything unless you can prove it beyond all doubt and intend to pivot to motivational speaking for the rest of your career. Outside of the defense industry (the DoD does pay generously to anyone tattling on Snowden/Teixera types), there is literally no incentive to speak out, and a lot of potential retribution you invite.
I don't get this automatic assumption that everyone is a hypocrite and no one is willing to face accusations of doing things wrong in their own life. I mean, is that how you operate? That's not how I do things. If I'm doing something wrong I love finding out about it - otherwise, how will I eve do things right?
if your goal for relationships with other people is to prevent them from discovering dirt about you then... I dunno man, I'm not sure how it's possible to live happily around other human beings with that kind of attitude. what kind of person does that make you?
> If I'm doing something wrong I love finding out about it - otherwise, how will I eve do things right?
We're talking about a few different things-- most of what I said applies to employer/employee relationships. If you hire a known whistleblower, you better make sure there are no skeletons in your closet because they've already proven what they're willing to do with such information.
> if your goal for relationships with other people is to prevent them from discovering dirt about you then... I dunno man, I'm not sure how it's possible to live happily around other human beings with that kind of attitude. what kind of person does that make you?
In both cases, it's just a matter of discretion. There is no prevention; discovery is inevitable.
okay but whistleblowing is good - I still don't get why you keep harping on this idea that if they revealed something secret about someone else, they'll reveal something secret about you, like - what kind of threat is that supposed to be?
again, I feel like this is you projecting - what are you hiding, anyway?
Yeah I find it a bit hard to believe. If you really think reviews are inauthentic it’s not hard to click on their profiles and see. I find it hard to see much evidence for astroturfing. If anything I think it’s fairly generous about not putting negative feedback front-and-center.
That, and, as the article details, and I’ve seen in my own personal experience, business owners are extremely aggressive about trying to get you to take down negative reviews, which I’m sure not only gets people to take them down but also discourages people from writing them in the first place.
I am friends with 3 different businesses in my town. A baker, a cafe, and a small restaurant. All three of them have experienced the exact same thing. I've seen it happen first-hand on their accounts.
If this is sufficiently widespread, seems like there's an opportunity for a honeypot operation.
I.e.:
(1) Guess at what information triggers this kind of shakedown by Yelp for a new business. E.g., credit-card sales data for new restaurants.
(2) Set up a fake business that pushes all of those buttons. E.g., register a new restaurant business in the state; temporarily rent a space and maybe put up signage; generate realistic credit-card purchase data and/or cell-location data for fake customers.
(3) Wait for the Yelp reviews to come in. [Possibly legal offence #1]
(4) Wait for the Yelp sales call. [Possibly legal offence #2]
(5) Turn down the sales offer.
(6) Wait for the negative reviews. [Possibly legal offence #3]
(7) Ask prosecutors to pursue criminal charges, and/or file a civil suit. Either way, get to the point of legally compelled discovery.
I wonder if this would be a fun first project for a newly minted government prosecutor.
> which I’m sure not only gets people to take them down but also discourages people from writing them in the first place.
I leave negative reviews for places that deserve them, and there's nothing they could ever do to get me to take it down, or discourage me from doing it. I doubt they could ever do anything that would make most people feel differently from me about that. In fact, I'd be pissed and just start leaving more negative reviews if they even thought of getting aggressive.
Well, sure, some will react that way. But some guy kept calling my wife and trying to intimidate her over my review. That probably works at least some of the time. And how many people want to go to court to defend themselves over a review? Seems like it’s not worth the effort.
Oh, I thought you were talking about anonymous reviews, where the the businesses can respond but not contact me directly. I wouldn't leave a review where they would be able to contact me. And even if they did, I'd handle that scenario a bit differently. Sorry that happened to you.
Could be individual salespeople doing it, and probably without Yelp truly "knowing". I worked at a nationwide gym once upon a time, and it was common for salespeople to trick people into signing long term contracts.
This theory is better than most, considering that other organizations have used similar tactics to get the benefits of bad behavior while dodging responsibility, such as HomeVestors of America (aka "We Buy Ugly Houses"):
Amazon also somewhat infamously uses contracting organizations with scandalous labor practices.
But to the contrary, these allegations have been levied against Yelp for nearly a decade, and yet no real clear evidence of this happening deliberately at least once has appeared. Meanwhile, people have switched to Google Reviews, which also have a pay-to-play element (for visibility) and dubious corporate backing.
One rather mundane explanation might be that people who click on restaurants near the top of the search results are just friendlier reviewers overall, while people who search to the fifth page are more picky.
If they go to the media, even if they are 100% on the whistleblower's side, it'll be a week (at most) of slightly negative PR. People will forget, if they noticed in the first place.
Is there a government body that will act on this? What will they do? A small fine? Would this actually go to court?
I worked at Yelp in 2018, and the allegations were there, I would be extremely surprised that a secret like this could be kept if it were true. It was generally a pretty open culture which lots internal open cynicism. There was just 0 evidence or even rumors internally this is true. I believe it's way more likely that the system is large enough to generate outliers that believe they're being targeted.
However, hyper aggressive sales reps were there and while I don't have specific details, I know their methods were the cause of at least some people not comfortable staying there. They however would not have access to things like reviews.
Is it possible that sales reps were using the publication of negative reviews by others as a tip-off of which businesses would then be more likely to purchase the service by which negative reviews could be removed? If so, the timing of those sales pitches along with negative reviews could make it seem like this is what was happening.
I mean, this isn’t the CIA here. It’s a tech company people are probably filing in and out of in about two years. And the salesmen are all meant to be in on it?
Yeah, if pay incentives are there, and we're talking sales... They're not necessarily people of the highest esteem. And if the hiring party puts a little pressure on the candidate, I could definitely imagine a tight ship.
A scrupulous person wouldn't join the Gestapo, y'know?
Most tech companies leak like sieves and the incentive isn’t there after you’re no longer being paid. It seems hard to believe it could be that airtight to me.
I dunno, man. I've seen lots of FRA, OSHA, MSHA, corporate, and food safety violations over the years - but they benefit me and my peers because they're shortcuts to the metric target. It's a win-win and everybody keeps it quiet, and everyone is on to it. And these are businesses that aren't selecting for privity.
I don't think Yelp's practices constitute a crime in the USA. The crime of a hypothetical Mafia is when they threaten violence for not paying. Yelp just changes what their site says about your business, which isn't a crime unless you can prove slander (very difficult).
Spreading falsehoods is libel. Pretending they can't know the truth doesn't shield them from culpability. Especially when their entire business depends on suppressing negative reviews.
It's possible the sketchiness is caused by a restaurant's competitors (writing negative reviews, falsely flagging positive reviews, spamming) rather than Yelp proper.
That adds a bit more complexity to the problem, but it still remains that Yelp is running a platform that to some extent relies on the ability to abuse it.
I’ve heard this story more than once from business owners.
Small biz across the world are subject to the reach of social media addicted primates with persecution complex. Few will take the risk of a review bomb or nobodies traveling to vandalize their business.
I have seen this same claim so many times over the years and not once I have seen direct evidence of it, just hearsay. "My friend", "My wife's cousin", etc. How hard would it be to show before/after screenshots, timestamps of Yelp sales calls, etc.
I would totally believe this. Probably not THAT hard to pose as a yelp salesmen and try to trick restaraunt owners into paying you. And of course the scammers could be quite good at gaming the review system.
This has been known for like 10 years now. I can't believe anyone actually still uses or trusts Yelp reviews.
There were lawsuits in the early 2010s over trying to reveal the identity of negative reviewers because so many restaurants were claiming they were fake/extortion.
Known to whom? You might be a sophisticated consumer but go hang out at the DMV and ask yourself how many of your fellow countrymen would really question yelp reviews.
which ended with:
The case was dismissed on appeal on the grounds that the complainants had not adequately proven (1) Yelp had wrongfully threatened economic loss by manipulating user reviews, (2) it was unlawful for Yelp to post and sequence both positive and negative reviews, (3) Yelp authored negative reviews, and (4) Yelp’s conduct amounted to a violation of antitrust laws by threatening or harming competition.
Yelp was also sued for false advertising (essentially saying they were a trusted, unbiased review source) which they somehow also won:
It's section 230 not article 230 but yes it's somewhat protected. You could go after the end user for defamation and you could probably use the legal system to force yelp to turn over identifying information from their logs if a judge agreed a comment was defamatory. But you'd have to go after the poster not after Yelp.
Just to be clear did you change from the default “yelp sort” to one of the other options? I don’t know many small business owners but the one I did know that didn’t pay Yelp had their default “yelp sort” results filled with negative reviews. Sorting by most recent the reviews were all positive and the one I left finally showed up.
It is easy to imagine that (some) Yelp sales reps are following an unofficial playbook to manufacture negative reviews around the time they engage prospects.
If one sales rep is caught, easy for the company to label them a rogue, bad actor.
A relative of mine owns an Air company that is as transparent and honest as they come (they've sent technicians who have no idea I'm related to the owner) and he was telling me they were a scam, but I had no idea that is how they work. I wonder how long until someone sues them for extortion.
If you think it’s bad now (spoiler: it is worse than you think), this is merely the calm before the storm. We haven’t yet seen the massive saturation of LLM generated content on everything from reviews to job applications to anything else you can possibly think of. The friction of sentiment-hacking operations will be dramatically reduced and we don’t have any effective mitigations. To make things worse, the only ones who can integrate the solutions (the platform owners) are not incentivized to take it very seriously.
The entire fake-review cottage industry harkens back to web search before Google was a thing. Search results were being gamed as implied relevance, ultimately making Yahoo and Excite and Alta Vista also-rans.
Online reviews need their Google-search moment. Something that levels the playing field, establishes a mechanism for credibility, and doesn't require an extortion-based business model.
The original model was as a long-form instagram live stream which I understand nobody did but I was being experimental. Basically a single shot, single take show on alternative platforms. The portrait format was to accommodate that media expectation.
I made dozens of them ... they failed to gain traction. A few are over 100 views a couple years later (like this: https://www.youtube.com/watch?v=3byGVjPq468) but I have still failed to crack the world of social media. It's not zero ... that's nice I guess.
I'm planning to do a reboot sometime soon. I took a couple weeklong trips around the country to some research libraries in the past year or so, but I'd like to execute differently to try to be more effective. This was my third show run attempt.
I think I may need a new foil for my host. The guy I did the shows with is a longtime TV comedy writer (https://www.imdb.com/name/nm1104065/) but he's not exactly a second city graduate.
The article links to one of her videos where she talks about the owner of a veterinary practice who buys reviews. I brought one of our cats there twice last year for after-hours urgent care, and my experience was -- fortunately -- positive both times. Friends bring their dog there for primary care regularly.
I'm just really wondering why the owner feels the need to do this. Is it a situation where you just have to do it, else you'll get buried under your competitors? Should I expect that our primary vet does this too, and that the other vet practices in our area also do it? Shady either way, but could it be a damned if you do, damned if you don't type thing?
Certainly there are business sectors where there's so much competition that some of them will resort to shady tactics like this to gain customers. But if anything, there's a shortage of vet practices in San Francisco (and the entire bay area, really). Gaining customers here feels like it's more about having available appointments in a reasonable time frame than anything else.
"I'm just really wondering why the owner feels the need to do this."
Everyone knows people who had negative reviewers are more likely to leave reviews, but what people don't often understand is, percentage of people who actually leave ANY review is still exceedingly small. I've heard as long as 1 in a few hundred or even a thousand.
So getting that first few positive ones is critical, to kick off the virtuous cycle of customer acquisition. You'd better deliver though, once customers start coming in.
It could just be the recommended marketing practice by the local small business school. Right after "Rent a week on the ad spot by the bus station" could be "buy a few fake reviews on yelp".
Reviews are just text associated with a date and a place. Seems like someone should be able to build a simple app or website that the public can use and trust for reviews, that doesn't cost much to run. We don't need huge businesses liek TripAdvisor (which mysteriously doesn't let you sort results any more). There's a few challenges. Fake reviews is one. Also the self-selection bias - people generally only feel compelled to leave a review after a bad experience. Or, being supportive, will be overly generous with the rating.
Perhaps if the review site could somehow prompt recent patrons of a set of establishments to not rate but rank the establishments, then some algorithm could convert that into a global score/ranking. You, an otherwise disinterested citizen, receive an email or text asking you to rank restaurants A, B, C which you have recently visited. You are incentivised to respond by a prize draw or something. Would this work? The hard part I suppose is getting the contact details of the potential reviewers.
I don’t use the reviews, just the listings. One of my favorite local restaurants purposefully asks for one star reviews on yelp, as they were listed there without their consent.
First and only Yelp review I ever left was three stars for an Indian restaurant in 2011. It was quickly hidden because "it didn't pass the bot filter" or something. I've never used it since as a reviewer or to find places. It lost all credibility on my first interaction with it.
My wife left a bad review for a restaurant and was stalked and harassed by the restaurant owner for years afterwards.
Since she couldn't get Yelp to help with that, it makes every review suspect. It doesn't take much of that behavior to convince people to only leave good reviews, and then they're basically useless.
I had a restaurant owner out me by name in response to a Yelp review. I pointed out that her attitude toward customers like me tells the world everything they need to know. That place went out of business, guess it wasn't just me.
How did they know your real name? Because they had your credit card and remembered you from your visit? Whenever I have a bad review to leave, I always wait a couple weeks, to minimize the chance they'll remember me...
> My wife left a bad review for a restaurant and was stalked and harassed by the restaurant owner for years afterwards.
There are multiple comments like this and I wonder how does it work? Like, you can just tell they guy to f-off, block his number and warn him that you are calling police if he continues to "stalk" you, no?
Yeah I had a similar experience with a plumbing business I chose not to hire after their rude and aggressive salesman came and insulted me in some misguided bid to get me to sign a contract.
I probably would have reacted the same way. But at the same time, Yelp's algorithm probably treats accounts with a track record as more legit than ones that were just created and then wrote a review. I wonder if they ever would have unhid your review if you had kept reviewing, or even just using the app to search for places to go.
Sort of beside the point, but I wanted to say how much I love the subject's whole vibe (to say nothing of her excellent mission). Her home and attire give me strong nostalgic feelings of late-80s after-church events in the SF Bay Area.
I keep seeing people swear by the NY Times' Wirecutter and they do have a much more upfront set of standards for reviewing products, but I've actually been pretty underwhelmed by the recommended buys from them. Still, I respect that they're trying. Wish that Consumer Reports was more easily available now too.
My unscientific feeling is that Wirecutter's gotten worse: less rigorous testing (particularly for reliability), more arbitrary reasoning behind their choices and what things are dealbreakers, more products excluded. A lot of things they seem to basically just pick some popular brands, talk about Amazon reviews, etc.
They also seem to have a much broader range of things covered now, so I think the strategy was to go cheaper, faster, wider.
Most review sites have a profit incentive that doesn't care about the quality of review. But I'll call out a couple that put in effort to be trusted.
Consumer Reports does independent testing with their established criteria. But it's a paid service and doesn't do user submitted reviews.
Steam gets gamed with review bombs but provides plenty of transparency and data for people to make informed decisions. They identify recent review biases and show spikes in volume. Importantly Steam has a generous refund policy with little friction, so they have a profit incentive to match you to a game you want to keep.
Not that I'm aware of. Reddit/HN has been compromised for years but I guess it's better than the alternatives. Whoever solves trustworthy, honest reviews will be a very, very rich person.
We will never have completely trustworthy user reviews. Reddit/HN quality is probably about as good as it gets.
Many small businesses get their viral loop started from glowing reviews from family and friends. That sort of growth hack will never disappear, and the subtle cases will be undetectable. I decline to write such reviews, but others won't.
> > Whoever solves trustworthy, honest reviews will be a very, very rich person.
> We will never have completely trustworthy user reviews.
just reading these to comments helped me figure out how to solve this: since apps have location tracking, for restaurants at least, your reviews should be weighted by how much time you spend close to a restaurant. If you claim to like a restaurant, but you don't eat there much, that means your reviews are not reliable. This unreliability would transfer to reviews of places that are not close to your home/work.
bonus points to combine it with confirmed food expenditures on your credit card/applepay.
It's not that people are making up their restaurant visits.
It's that they are going to the restaurant, they had a good time, but they didn't disclose that they are close to the owner, and maybe they overlooked a few things here and there that an impartial reviewer would note.
Anything that works gets gamed, the only real possibility would be various sites that are so absolutely local that attempts to game them would be painfully obvious.
> It's amazing to me how many people have conspiracy theories about Yelp reviews, when in reality it's just really hard to determine good vs bad actors in review systems and it's possible to do a bad job as an engineer.
When cards are cloned, it's exceptionally easy for a bank to determine which merchant is compromised by looking for common denominators.
Take three actors.
What are the odds that two of them have reviewed one business, and only that business? Pretty high. All good. Same for all three.
Now, what are the odds that two of them have only reviewed two businesses, and both businesses are the same, on both accounts? Still decent, sure. All three actors, same two businesses reviewed, and only them? Getting a little dicier.
Two actors, seemingly independent, have each reviewed three businesses only. And they've reviewed the same three businesses, in the same order. And perhaps they each reviewed the first business, then two weeks later, the second business, and then the third. Odds? Getting pretty slim now. A third actor, doing the same? You're stretching the bounds of credulity for sincere reviews.
This stuff is basic probability theory. Looking across reviews for sets and unions of reviews and reviewers, and increasing the number of reviews, but still seeing commonality between them points to SOME malfeasance worthy of review.
This comment is fairly dismissive of GP's experience and intelligence. If you think you can solve a problem just by spitballing it, and GP's team at yelp couldn't figure it out for years, either you're much smarter than everyone on their team or you're underestimating the problem.
I worked on an anti-fraud team in a similar role, and I agree with GP that it's a harder problem than people realize. It's easy to come up with a simple metric that would identify the fraud that you observe, but you don't know what false positives it would catch as collateral damage.
Like this example that you find wildly implausible:
>Two actors, seemingly independent, have each reviewed three businesses only. And they've reviewed the same three businesses, in the same order. And perhaps they each reviewed the first business, then two weeks later, the second business, and then the third. Odds? Getting pretty slim now. A third actor, doing the same? You're stretching the bounds of credulity for sincere reviews.
I think it's highly likely that if you used this metric to find patterns, you'd find fraud, but you'd also find lots of legitimate reviews (e.g., friends who go to the same restaurants together, friends who recommend restaurants to each other).
> This comment is fairly dismissive of GP's experience and intelligence.
I don't have the full text of that comment, as they've edited it our, so I don't know specifically what they said. But perhaps they just don't know everything about how Yelp operates; it's not a small company where everyone knows everything.
I also know someone who previously worked at Yelp, and they've talked about some of Yelp's shady internal practices. Everything is not all above-board there.
I'm willing to bet there are more people who could solve that problem working for yelp than you would fairly estimate there are, but the suits are actively preventing solving that problem because they know that shaking down businesses for Yelp support contracts is infinitely more profitable than being a reliable source of truth in the world will ever be.
Gold Plated Bugatti's do not buy themselves, after all.
That's your perspective. That's when -your- behavior doesn't match your previous behavior.
What they are looking for in a different context is "Customer A reported unauthorized transactions, and so did Customers B through D. Each of these customers, prior to these reports, all had a transaction occur at XYZ gas station, so is there a skimming problem at that gas station?"
I can see some ways that might go awry (what if I like reviewing every new restaurant that opens?). But more importantly, bad actors will figure out and work around each heuristic like this you make.
Another way of saying what I’m trying to say is that at any given time, no matter how much they keep heaping on new heuristics, there will be some missed cases that should be obvious to a human observer.
It doesn't matter what you did as an engineer, and in fact automation isnt really a big part of the discussion.
If sales people etc have special access to add or remove reviews, which they do, or at a minimum leave negative reviews themselves when you don't pay, it is corrupt. It is well documented at this point that they do this.
I would believe individual salespeople might do unscrupulous things like enlist their friends to leave bad reviews or make promises they can't keep about altering someone's score but that's a totally different accusation than saying Yelp as a company has a policy of doing this.
1. The ones unable to effectively police shills and spam. I'd put Google in that bucket.
2. Those that turn a wilful blind eye to review fraud, as long as they boost sales, like Amazon
3. Those like Yelp or Glassdoor whose entire business model is built around extortion and a protection racket ("nice business you have here, it would be a shame if a bad review happened to it. Unless you buy ads with us, of course").
> She also reached out to Sasha Ganji, the lawyer whose signature was appended to the letter she’d received from the law firm appearing to represent SavantCare. Ganji never touched the document, he told Dean; apparently someone had used his name and signature without his knowledge.
That is a serious thing to do. This is criminal fraud right?
I don't work at Yelp though I'm pretty sure that the reviews which stay on the website for any length of time are those posted by users who actively use the website to search for businesses (i.e., tracked via their cookie)
i.e., while there certainly can be paid/manipulated reviews, the barrier to doing so is higher here than elsewhere (e.g., google reviews, tripadvisor, etc.) and there's enough people who are happy to use other less controlled sites, I haven't witnessed any truly widespread fraudulent reviews like I have elsewhere)
Generally speaking, I'd say Yelp reviews are, generally speaking, comparatively the most reliable
I can't speak to how they sort/present search results - that does seem potentially shady
I worked on a start-up that was basically a Yelp competitor with a less-bad business model. Unfortunately, all the incentives in the business are about boosting the performance of your paying customers relative to free businesses on the platform. While we never went as far as faking reviews or helping businesses eliminate bad reviews and we at least were clear to customers about our search lists being curated rather than complete, there were clear pressures towards all the things that lead to Yelp being the shit-show that it is.
What astonishes me about both Yelp! and Trip Advisor is their complete inability to grow their share price, even with near-monopolies in Restaurants (Yelp!) and Hotels (TripAdvisor). Yelp! catalogs 205 Million reviews last year. 30 Million for Trip Advisor.
But even with such dominance, in the last 10 years, Yelp! is up about 38% and Trip Advisor is down 42%. What is it about their businesses that makes profitability growth so elusive?
I hate to be dismissive, but this is kind of a "Water Is Wet" article. Does anyone really think online reviews are anything but totally gamed, bought, and paid for? It's kind of a Universal Law Of Spam: Any place on the Internet where users can add content and where content might be profitable to someone, will eventually end up full of spam and paid-for content.
Of course some reviews are real, and there's an art to deciphering the wheat from the chaff. The problem is that there's the constant pressure you describe to drown out legitimate reviews or otherwise game the system, so that 'art' becomes more and more tricky and less reliable. I gave up on reading the Yelp tea leaves a long time ago, and other prominent ones like Google Maps and Tripadvisor have been getting worse over time. I can still usually intuit my way through Amazon and and Steam reviews, if there's enough reviews and I take my time.
There's huge money to be made pretending spam isn't a real problem. It used to be that spam was universally despised, fought and even prosecuted in some jurisdictions. But it turns out that it's cheaper (and actually profitable) to join the spammers than to fight them, especially if you have an endless supply of marks that will guzzle the spam without second thoughts.
Funny to find your post at the bottom of the thread.
I tried to use a few apps for finding restaurants, compared their rating with the experience I had and concluded it's a waste of time and money.
There's not even a need to know or understand anything about it. No insider info from Yelp, not experiencing it as a business owner. Just try and see for yourself.
When I was on a trip in the US for vacation, I found out that most of the review services can’t be trusted. Despite good or bad reviews it was a complete lottery. Meanwhile in Poland when a restaurant has a good review online you can actually expect reasonable service and food quality.
The Apple Maps integration seems to be most of what's propping them up these days. It's especially infuriating because you can only see about 2 reviews before you're prompted to download the Yelp app for "full access".
I thought Apple was supposed to be working on their own review system or switching to a different service, but I haven't seen anything come of that.
I use Apple Maps and do have some complaints, but I like when it tells you which lane(s) to be in as you approach freeway interchanges. Sometimes it's obvious where you're supposed to be (far right), but sometimes when consecutive exits are close together, you actually need to be 1 or 2 lanes over.
I find that if a restaurant has 4.5 or 5 stars (assuming a decent number of reviews), typically it is very good. However, plenty of the places I eat/enjoy have 3 stars (none have 2 or 1). So I use it as a rough filter for that purpose.
It's also useful for seeing if particular menu items are good, since fake reviews don't typically get into that much detail (or at least there are other real reviews that would outweigh them in search results for that item).
There's the ongoing meme that the best Chinese restaurants have 3.5 stars on Yelp- from the combination of 5 star reviews from people who love the food, and 1 star from people who didn't like the brusque service or didn't get the American Chinese food they were expecting.
There's a Malaysian/Chinese resto near me like that. It's an absolute sleeper of a place to have eat. Barely visible from the road, and connected to a supermarket (and they wheel in and out shopping carts from next door as they run out of ingredients during the day). My Japanese roommate at the time was turned onto it by his Korean boss, and I took my Chinese/Filipino/Indian/American coworkers there and then they started going there all the time.
The 1 star reviews complain about the organ meat, which there were totally grossed out by. "Beef tendon pho" is super legit, and what you get in Asia. You're in the wrong place if you expect American-Chinese fare.
It's super weird to me that people will leave a bad review because there were things on the menu they didn't like. I like some types of organ meat, and dislike others, but it would never even enter my mind to be like "2 stars because 40% of the menu is stuff I don't like". People are weird.
I have another comment in the thread which got downvoted by someone.
My take, having seen a lot of these posts on yelp, is that people leave such reviews because restaurants are offering something which they are not familiar with, or does not match their notions of "normal" (however narrow that may be). You see this less at restos which serve burgers and pizza, and more with places serving non-American food.
I just took a spin through the couple of Asian places in my town, and they have relatively high numbers of 1 star reviews. Digging into them, what you see a lot of are is complaints like:
"This is the worst Asian food I have ever had because (proceeds to describe exactly how the dish is supposed to be)". Also a lot of complaints about these Asian restaurants being "dirty" (which they aren't - as I'm only reading through places my wife and I go to / order from frequently). It was like this specifically with Asian food in the last town I lived in as well. I recall one complaint about how they romanized a Thai word, insisting "pad see ewe" was misspelled (ORLY!). It's bizarre. Mexican restaurants by comparison, get complaints about the food being too spicy, or the decor (at takeout places no less) being bad. And, of course, "dirty", which again, the places I frequent, aren't.
I can't help but conclude that people who write these reviews are judging things based on their own biases, of which they are not self-aware. A more harmless example is a review from a great Thai/Chinese resto where I used to live where someone suggested the place smelled "musty". Well - it was in a sublevel of a large building ("basement" is the word they used), but it was not in any way musty in the 10 years I spent going there. None of my friends ever took issue, even a good friend with a really sensitive nose.
It surprisingly has up to date content where I live, and generally anything 4+ stars is good and below 3 stars is not too good. The individual reviews are usually not useful, but looking at patterns and trends of recent reviews can be useful if you know how to filter out the obvious BS and one-off overreactions.
The problem is a real lack of reviews for smaller places, especially outside the downtown area. Google Maps has more content there.
I agree, though the other issue for smaller places is that they are more susceptible to review-bombing by one disgruntled party (do we really need 6 reviews from 6 different people about how the same table had to wait an hour to get the check one time?) or "5 stars for 30% off" incentives (which I have unfortunately seen a number of, especially at small businesses trying to distinguish themselves).
Google Maps I still don't understand where they get a lot of reviews from in the first place. While some look like normal reviews, there's a lot that have no comments at all or comments completely unrelated to the place ostensibly being reviewed (or just cryptic things like "not good" or "good"). It's not a bad comparison to use alongside Yelp, but it's definitely led to more clunkers, especially for places with n<20 reviews.
At least in New York, it is. It’s less prone to the cost noise that infects the other rating services—where a 3 star could be due to bad food or someone being surprised by New York prices. (There is also a scourge of paid top ten articles that makes a simple search useless.)
I don't use Yelp to determine if a place is good or not.
I use yelp for personal entertainment, because the outraged, bombastic reviews are often hilarious in how just beneath the surface are classist, racist, xenophobic and jingoistic undertones. These are among the few reviews I believe 100% are actually real people!
I personally prefer more objective signals like the menu, pictures of the food, and pictures of the interior. Maybe I’m a snob but foods that people describe as “delicious” and “amazing” to me look greasy, salty, or altogether not much better than I could cook myself with basic cooking skills.
With the average palette you just need to sprinkle the right amount of salt and sugar to get "yummy" reviews. These people can't cook, barely know where food comes from, and have no way to judge food other than it being sweet (good) or not sweet (bad).
I forgot it exists. Looking at a bunch of local restaurants, there has been essentially no activity since 2017, and Google and Facebook have orders of magnitude more reviews in every case.
Online reviews have become one of the dankest cesspits of misinformation and deceit on the entire internet. Amazon's reviews are so thoroughly gamed as to be functionally useless and Amazon knows this and tacitly supports it through their inaction. Yelp and Google Places review are being gamed en-masse by bad actors and scammers and have become so distorted that by relying upon them you're now more likely have a bad experience than a good one. Review websites are stuffed full of advertorials and universally have such super shady advertising / sponsorship policies it's functionally impossible to know whether the review you're reading was paid for or not, and all the big tech sites are so dependent on advertising revenue, they almost never post critical review for fear of upsetting their advertisers. It's a unspoken rule in the industry these days that negative reviews are simply not published. Not to mention damn near every single product review video on YouTube, IG Reels, Tiktok, or wherever else now is either straight up sponsored content where the "Reviewer" is receiving monetary compensation for their so called "Review" or at minimum is receiving the product for free, which creates yet another perverse incentive, as anyone who posts a negative review will simply stop getting free products to review in the future.
There are literally dozens of 100+ person companies in China and India who's entire business model is faking online reviews.
The entire industry is broken, and the average consumer can't find literally a single source for unbiased, trust-worthy reviews of anything anymore, which I for one believe is the system working exactly as designed and intended. Never forget the rich will sacrifice literally anything that's generally considered to be part of the "Public good" if it means they'll get rich faster, which is exactly what's happening here.
Go Kay Dean! Yelp should hire her and give her a team. It's a win-win. She is clearly passionate about the problem, and effective. Yelp gains added trust of their bread and butter, the review system.
I assumed Yelp reviews can't be trusted from day one. It's obviously a huge target for gaming and scamming, pretty much any review site is. I include Angie, and similar sites in that as well.
What’s the solution then, without the need to trust centralized companies? I can’t think anything besides system with a WoT (web-of-trust) architecture.
5-star, less than 100 reviews: probably paid for reviews. I've never been to a "bad" place in this category, but I usually avoid it unless it's the most appealing option, so my sample size is small
4+ star, 1000+ reviews: definitely not bad. i would rate every place ive been to in this category as good or excellent, although some are overpriced. all of my favorite places in the bay area fall between 4.0 and 4.5
3.5 star, 100+ reviews: possibly good, probably not bad. ill check reviews, if people are primarily complaining about service and not food, it's probably good. i never pick 3.5 places if it's somewhere i've never been before, but i won't veto one if someone else suggests it
3.5 star, less than 100 reviews: i avoid these completely. i've still never been to one
less than 3.5 stars, any number reviews: i also avoid these completely
So many people see the brewing storm of AI as dystopian, but this is one of the tasks that it could be quite good at.
Humans are biased, deceptive creatures and whether intentional or not we can misrepresent ourselves, others and every...thing. AI would definitely help to drop a "badge of approval" on something in some cases.
Reviews for restaurants may be harder, but just simple fact checking it's going to be so amazing to have a trusted AI that can say "yep, this is most likely true" with a click-through to references/sources/reasons why.
Look at trending Twitter tags, I've been looking at #ukrainerussiawar etc a lot and sometimes like 90% of the tweets have _nothing_ to do with the topic, jsut shameless self-promotion spamming tags, or misinformation tweets from either Russia's information warfare team, or quite simply right wing morons parroting propaganda. Would be nice to make all of that go away...
Man I forgot about Yelp until I read this. Apparently its because its packaged inside Apple's "Walled Garden" Maps. Seems like they got a weed in there that needs to be picked.
i never use yelp, i can typically tell what a restaraunt is like by the name and pictures on google, and whether they serve craft beer or brand names, i don't look at reviews or how much it costs
I was glad to see that the horoscope was only the 7th most popular page on SFGate today.
I can only imagine that this means all these recent troubles I've been hearing about San Francisco have been resolved and people do not need the help from their horoscope anymore to know what to do to resolve their problems.
I am pretty sure that we can use clustering for that. As they said there was a clear pattern the restaurant recognized. If they had all the data it would be easy. The reviewer could also ask for a token from the restaurant before submitting the review or a signature, there are plenty of options.
In the future, reviews will be automated, thus immunizing them from unsavory practices such as paid reviews. Your GPS-enabled wearable compute device will determine which establishment you patronized and for how long, and biometric data will determine whether you enjoyed your time there. These data will be "anonymized" and aggregated. "90% of [brand] Watch wearers like You experienced heightened happiness scores at this restaurant!" Trust in reviews will have been regained and the world will be a better place.
Hello, [restaurant name], this is Alice from [brand] calling. Would you be interested in our premium marketing package? Features include enhanced filtering of "anomalous" (negative) happiness-score(tm) readings. The subscription costs only 1,337 credits/month.
Good point if the app was independent like Yelp. I envisioned this as a first party product from a more powerful company, which makes the Watch as well, that could then directly integrate the consumer/review data into their Maps app on their Phones and other devices.
Alice from [brand] would then have to find a way to get customers to take off their watches or somehow artificially inflate their happiness scores. Or, funnel naturally happier people to their establishment, similar to lead sourcing.
If I search for "dinner" on Yelp for "current location" and look at the map, it leaves out an astonishing number of places that I can see within a mile if I drive down the street.
And if your business model is centered around hiding all but the winners - who might pay for placement + reviews both - then you have to try to keep people from talking about it, and the moral interests just get worse and worse.
Google Maps isn't any better; OpenTable seems the "least bad" of options I've looked at recently, at least for fancier reservation-having places, in terms of actually showing my options.
(It's a broader internet problem in general - "trending" algorithmic reinforcement stuff will push everyone towards the ONE TRUE BEST HOTTEST MOST POPULAR bbq/ramen/whatver place, when probably there's half a dozen other options within 10% of taste and quality, that some people very well could legitimately like better if they heard about them too.)