[jefftk]

I can't speak for Google, and I don't work in that part of ads, but my understanding is it's primarily because of how adversarial fraud detection is. If you give people who violate policy detailed information about what they did wrong and how you know then it is much easier for them to figure out how to abuse the system without getting caught. I don't like it, but I also don't know how to fix it.

[grog454]

>If you give people who violate policy detailed information about what they did wrong and how you know then it is much easier for them to figure out how to abuse the system without getting caught.

I don't really buy this.

The most effective adversaries already have a deep understanding of detection mechanisms and are typically just tweaking parameters to find thresholds of detection. Other companies mitigate this by delaying bans and doing "ban waves", or even randomizing the thresholds (I have done both for certain types of automated bans for attacks on my systems).

More to the point, adversaries already know what they did wrong so telling them isn't going to make much difference. False positives do not know what they did wrong and telling them will make a tremendous difference.

Full disclosure: I have been the victim of a false positive flag in that my app with over 50M downloads on google play was removed and then reinstated when my reddit complaint post got human attention (thank g̶o̶d Google).

[pdonis]

> adversaries already know what they did wrong so telling them isn't going to make much difference

I'm not sure that's true. Adversaries know they are intentionally trying to game Google's system, but that is not the same as knowing all of the internal parameters of Google's system. Telling them what they did wrong in specific cases might well give them useful additional information about those internal parameters that Google does not want to give them.

> False positives do not know what they did wrong and telling them will make a tremendous difference.

While this is true, it is also not actionable, since the whole point is that Google does not know which positives are false positives and has no cost-effective way of finding out (since finding out would require actual humans and the scale of its ad business is too large to make the number of humans that would be required affordable).

[sideshowb]

I don't for a moment believe Google couldn't afford to pay more humans. Do you?

[pdonis]

Not the number of humans it would take to replace their automated fraud detection algorithms to dramatically reduce false positives at scale, no. The whole point of their business model is that all of those processes have to be automated, otherwise they aren't profitable.

[saltdoo]

I had a similar experience and haven't seen it put this way before. Great point about false positives.

[ent101]

Thank you for the response. Please read this comment.

This does make sense, but why the atrocious support? I mean at least reply to emails even if we have to wait for a month to get a response, just give us something real not the canned, automated email.

I've never ever clicked on any of my own ads EVER and I got suspended for "click fraud" and I'm sure there are many people like me... Why do I have to be terminated without any recourse? This feels, for the lack of a better word, pure evil and cruel. This makes people deeply hurt and hate the brand.

I'm a developer and a business owner and because of my experience with Adsense and Adwords there is NO WAY I'll ever use Google Cloud, no matter how much discount Google is willing to give me and I'm definitely not alone in this. This behaviour is going to destroy Google in the long term. Right now you guys make money on advertising, but as soon as you need to sell something, I'm not sure people who have been hurt are willing to pay for it.

[js4ever]

Agreed, I have been burned in the past (2015) by a ban on google play store and adsense on one of my business. It was impossible to talk to a human or get any details and this killed the company.

I'm now a Google hater, I'll never do any business with them like EVER even if they pay me to use their products.

Also I'm the opposite of an ambassador for Google brand. I managed to turn several huge customers in EU and US to AWS or Azure solutions instead of Google cloud... Several times and I'll continue to recommend companies to not do anything with Google. Yes in the long run their attitude and people's like me will continue to destroy their brand.

[anaccountexists]

I work with a lot of fraud teams at a FinTech company, and we have the same exact problem. If you give out too much information around what caused the enforcement action, people are able to figure out the exact formula used to trigger a risk check, and will float right under it.

It’s a really hard problem.

[raverbashing]

> I don't like it, but I also don't know how to fix it.

Google does not know how to fix this because they don't think outside of the tech box. This is very much an issue with their way of doing things.

You build relations with advertisers. If someone is spending 20k per day on your system (for months or even years), you better have someone to talk to and actually look at the issues they have.

"Do things that do not scale" as PG says.

[jessaustin]

In case you work in the relevant part of ads, could you say why advertisers are allowed to buy their competitors' literal trademarked business names on Adwords? This seems like a protection racket: "nice business you got here; it'd be a shame if that knockoff down the street paid me enough to bury it..."

[grapehut]

Absolutely agree. The company I worked from had a persistent problem with someone stealing customer details by buying AdWords targeting our brandname with a fake phishing site. We would report it, 2-3 days later Google would take it down. Then the next day, they'd be back with a different phishing site. At what point is Google just complicit?

[jessaustin]

Wow I didn't even think of that one.

I saw it when a business for which I did some work would have customers calling from their competitor's lobby saying, "This stupid receptionist doesn't know about my appointment, what do you mean I'm in the wrong place, I put your name in the phone google?! Screw you I'm going to just get the work done here!" Google/Maps/Android had customers driving past the business for which they had searched by literal business name to some other business that was spending more on Adwords for that literal business name.

[jessaustin]

By the complete lack of response from the Google representative above, one may infer that he works in exactly this relevant part of ads, and has nothing to say in his defense.

[pdonis]

> my understanding is it's primarily because of how adversarial fraud detection is.

Let me rephrase this to make it clearer what the actual issue is: it's primarily because the cost to Google of false negatives on fraud detection (failing to detect actual frauds) is much higher than the cost to Google of false positives on fraud detection (flagging users as fraudulent that actually aren't). So Google is willing to accept a large number of false positives in order to avoid false negatives.

Or, to put it more simply: the incentive structure of what Google has chosen as its core business model means it is in Google's interest to randomly penalize a large number of bona fide users simply because a much smaller number of users are fraudulent.

In other words, this is basically unfixable unless Google changes its core business model.

See the problem?

[protomyth]

I actually wish some Congress critter would propose a law that would require companies to "give people who violate policy detailed information about what they did wrong and how you know" and require a registered human agent for communications in these situations.

[datavirtue]

So the policy can't be found in the TOS and the reason is really a lie? Obfuscation is complicity--part of the problem. Get it all out in the open and the problem will resolve itself. The bedrock of our most hallowed institutions, and principles of democracy and free markets, supposedly refute your reasoning...if you believe in any of that. Listen to yourself as you apologize for the master scammers.