Cheap signs along the road don’t trip that heuristic because they cost so little it doesn’t change the underlying economics.
I was actually interested in some of those privacy/info removal services but after doing research found those to - as you said - lack value for the money.
Advertising intensely to us is the absolute best way to lose us as a customer.
However when you’re advertising a VPN on a cooking channel the cost per customer is quite high so they need to recuperate that high cost by charging extra. This is more true the longer the advertising campaign runs and the less a channel is related to the product, each of which drive up new customer acquisitions costs.
Obviously it’s not a perfect predictor, but it doesn’t need to be.
If person X says "ads don't work on me", the state "I experience no influence from ads because they don't work" is indistinguishable from "I experience no influence from ads because they're so sneaky that they only affect me subconsciously".
Unfortunately, it's very hard to get individual-level evidence. You can get population-level evidence, but sometimes that evidence shows that the ads don't actually work (for instance, The Correspondent's 2019 articles about the subject).
They have always had powerful psychological tools but they are next level nowadays. Best to just avoid.
* https://i.imgflip.com/2yg87r.png
(I don't think pokemon intentionally wants such a toxic secondary market tbf)
