Maybe off-topic, but I see there is a giant difference in how I feel regarding tv/radio/street advertisement vs Internet (mostly targeted) ads. In old media, there is a lot of variety in ads, to the point where basically no particular ad ever became too annoying to me (with few exceptions). And the specially cool ones give me good feelings.

But when it comes to Internet ads, I find it's the opposite. Take Youtube for example. It is often the case where Youtube will show me the same 2 or 3 ads over and over for quite a few weeks. That is extremely annoying, and makes me want to never ever have to deal with those brands. Seriously, fuck those two insurance companies, or that macho soap brand, also Google Pay, or that AOE clone game. It is extremely annoying. I know exposure to the brand leads you to become familiar to it so more likely to buy it, but the kind of extreme exposure that some internet (targeted) advertisement makes me associate those brands with negative feelings. That can't be good for their brand. </rant>

It's still very debatable whether advertising works at all, not just targeted but in general. And if it does, the extent is questionable too. Freakonomics had a very good episode on this a few months ago - https://freakonomics.com/podcast/advertising-part-2/

Not all advertising works. Obviously. That does not mean that advertising does not work or that it is debatable in general. It is debatable for specific scenarii. I think that episode of freaknomics says pretty much that.

What this means, IMHO, is that it is very important to be able to measure results so that you can decide based on data and stop throwing money in the wind if it in fact does not make a difference.

If you want to confirm something where others have an interest, you may easily find all sort of confirming results.

What you'd like to look for is those results proving it doesn't work. Personally, ads are simply an annoyance and sometimes even hostile at times.

Targeted advertising works and delivers better results than non-targeted advertising. This has existed for pretty much as long as advertising, way before the internet.

I am not expert in the field but I am sure that there are hard data from studies available on the subject.

A question is whether it is worth paying the extra money to run a specific campaign on a specific platform and the answer is, as often, "it depends".

> Targeted advertising works >> I am sure that there are hard data from studies available on the subject

Those are contradicting, but I suppose there is some A/B testing data somewhere to confirm that advertisement do work

They are not contradicting. I'm just saying that I don't have references to studies on hand, but I'm sure that they are available because, well, targeted advertising works (Hardly a ground breaking claim, that is pretty much established for a hundred years at least, and you can run your own campaigns to convince yourself).

I agree that targeted advertising worked to some degree, you wouldn't advertise a Rolex in a suburban housewives magazine, a luxury yacht in a gardening newspaper or a cheap alternative to butter in a cooking weekly.

Today's online targeting have two flavors AFAIK, one resembles the old way of putting an add in a place where we expect people to react to it. The other works differently, they try respond to search terms or site visits and my feeling is that this kinds doesn't work so well.