Thanks for the post, Nick.

I'm familiar with some issues related to bot behaviours (1) on Google Ads, and as far as I know, Google doesn't try very hard to filter or detect such bots — on some verticals they might be half of the traffic, and Google still earns from it.

But I found a potential solution in Google itself, since they provide a tool to select your demographics. If you select all available options and skip Unknown, it won't show the ads to some part of the bots. There will still be some percentage, but this cuts out the largest part, since most click bots have no Google history and therefore fall outside the demographic buckets.

1. https://www.tirreno.com/bat/2025-12-27

Bot farms may make up the bulk of "normal user behavior" at this point. It would explain at lot, actually.

I received a phishing email that appeared to come from Google, that almost had me convinced. I was in a delicate place--my father was fighting cancer, and my brother had just been horribly injured in a motorcycle accident. I was distressed, and my wits were not about me. The link in the email was to a google.com url, it looked okay.

Luckily I detected the treachery before sending any sensitive data, and I sent the relevant details to the Google bug bounty. Their response was that google.com open proxies are not considered bugs, and the report was closed. Do less evil, Google.

Google doesn't want to remove bots. It inflates their numbers. See how many devices saw your ad? Millions!

And yet they trap me in endless captchas because I use Firefox with an adblocker.

This is a really good marketing strategy to drive traffic.

Bots have probably existed on google for two decades, so probably a boiling frog situations.

For at least 25 years, bot farms have been an easy way for ad presenters and mediators to fraudulently profit without showing ads to real users. It costs money and effort to disambiguate humans from bots.

Congratulations, now you are an Google unpaid employee

[dead]