Is there a way to prevent gaming the system like this, perhaps other than whistleblower rewards? I imagine even with whistleblower rewards companies would just employ third-parties for plausible deniability claiming it's a simply the actions of a bad actor trying to get them penalized.

Edit to add: perhaps substantial whistleblower rewards for employees of companies doing the bad actions but the top one or two employees/founders might be the only ones in the know.

A whistleblower checkbox "seller had promised to compensate me for this review (this report won't be shown on the review or visible to the seller, and you won't be penalized for this)" may make an impact. Game the game. Collect those flags for a while without visibly affecting anything, if enough are collected - verify existence of those paid reviews (to avoid report abuse attempts to shut down legit sellers), and then take action.

Bad actors would probably start using it against competitors just like the fake 5 star review attacks. It's really interesting how seller competition evolves, anything that can be used to punish a bad seller ends up being used as a way to frame a competitor.

Yeah, it's why I ultimately think opt-in verified identity networks are how we counter that; not necessarily an easy task to educate people why, educate what's acceptable/unacceptable, and for them to trust the leadership/governance of that ecosystem to be willing to have a unique ID so a bad actor, even if simply.in a timeout for misbehaving, can't circumvent integrity systems by creating another account under fake or stolen identity; a cascading or hierarchical or decentralized-like system for different nodes of trust-identity verifiers may be useful or a good idea, or not, and then can have umbrella organizations that determine which subs or arguably smaller managed-curated networks are trustworthy with their processes/protocols - which can be "random spot tested" via umbrella organization's verification mechanisms to check for integrity; that's unnecessarily complex or it allows for different existing verification platforms to aggregate - but they may not be willing to integrate with such a system as attempt for competitive factor/leverage.

Sure, the very same day such feature would be available. That's why I explicitly mentioned need of independent verification. It's impossible to verify every seller out there, but individual sellers can be verified. Those tips must be merely signals to focus, never to penalize a seller solely based on the flags.

Also, the existence of the checkbox itself must have some chilling effects on the sellers by changing their risk perception. Those days it's just a common knowledge that those "leave us a review and we'll reward you with a $X gift card" is not penalized in practice. A checkbox with a PR campaign would send a contradicting signal, making this practice more scary.

A resourceful fraudulent sellers' cabal (if there are such things) could try to poison the well by paying users to tick this checkbox in random reviews (wasting marketplace's resources by muddying the waters), but this doesn't sound scalable and doesn't feel like a realistic scenario.