> 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.
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.