>curiously omits the details of why they were banned
That's a difficult thing to accomplish when Google has not provided them the details of why they were banned.
That's a difficult thing to accomplish when Google has not provided them the details of why they were banned.
It sounds like he has violated something multiple times and was told what was in violation but he is choosing not to share this part of the story
So it is not so clear if they received anything useful before or not. They did use ads and the ads team is quite trigger happy, which is interesting because competitors can indeed cancel your business by creating click fraud. The ‘AI’ can detect click fraud, but cannot filter it (wat?) so this happens a lot.
I helped someone out with a similar issue, a friend of mine who was doing something very very uncontroversial.
The "previous messages" had no extra information at all except, "You might have broken our terms of service with this page" - a link to their main page, with no additional information.
It doesn't, actually. Everything I've read about Google indicates they're extremely opaque and will ban for bizarre reasons.
Google itself might not even know. After all, machine learning isn't known for being able to explain itself.
> The ban letter says they previously received other notices. Where are they?
IIRC, I understand that someones those notices come all at once at the same time as the ban, if they come at all. It's also quite possible that the ban message from Google is in fact inaccurate.
> And what steps, if any, did they take to resolve then?
What steps could they take? My understanding is that Google offers no recourse to a human, and their automated appeals process might as well be a "reject all" rule (if it isn't literally that).
You seem to be operating under the assumption that Google has implemented a fair and reasonable process, so the onus is on the person who runs afoul of it. However, everything I've read has led me to believe Google doesn't have a fair and reasonable process, because they rely too-heavily on algorithms and automation.
Indeed; it is just multiple flagging by ML and that is it. Not much you or anyone can do.