Like TV. That way my data is safe, and I'm still exposed to your ads that are relevant to people visiting 'x' site.
The first one is useful for me, but the niche of people who would buy it is so small that it wouldn't make sense to target it to everyone. I hate the second ones, but the majority of people love buying cars.
I know a lot of very non-technical people, some of who dislike facebook in general, talk about how good facebook is at showing them ads that actually get them to discover and buy new products.
The people I've heard this sentiment from don't feel like they were manipulated into buying something that ultimately didn't meet expectations. I commonly hear things like "I can't believe how good the Facebook ads are. I only see ads for really nice Blouses in colors and sizes that I like. I've bought nice things via Facebook ads".
Granted, it's usually followed up by "I can't believe Facebook is able to be that good at showing me the right ads" ... but people I've talked to seem to feel neutral about the creepiness of it, I'm guessing because the upside is they get to discover and buy cool new things that they otherwise wouldn't have known about.
(Personally I don't like ad targeting on the basis of companies abusing / selling / harvesting data about me... but for the average person I think seeing well targeted ads is positive UX compared with showing non-targeted ads)
During my time at Apple one of the big things they impressed on new hires is that there are things we can do with technology, that may even be beneficial to the user, but will be perceived as creepy. For instance, Starbucks could use facial recognition to greet you by your name at a location you’ve never been to. They could have already started your drink with a high probability it is what you order based on time and weather. But these things are “creepy” to much of the population and should be avoided. Ad networks do not care because their customer is not the end-user.
AdSense then proceeded to push just that one "Pebble Smartwatch" ad on me, everywhere. Even a page with multiple ad blocks just had several of them -- sometimes all of them -- the same text ad for Pebble. It became more annoying than helpful because I was already going to buy one, it just tried to push me to buy it for a higher price.
I turned ad personalization off again and now any ads I get are mostly just related to the current page's content.
The first one I'm actually sympathetic to. The second I vehemently disagree with.
I don't think that anyone would disagree with you that true product discovery - educating you about a thing you didn't know about that solves a problem you have - is a benefit of good advertising. Lots of advertising isn't about product discovery, for sure, it's about persuading you to buy a thing you don't want and reminding you about a product you already know about and building an image about the product, and many people want to get off the hedonistic, materialistic treadmill that such persuasion pulls towards, but let's ignore that for now and assume that targeted ads lead to a benefit to the consumer of improved product discovery.
However, I don't think this is a very big benefit, in the real world. I can count maybe 3 things that I've bought in the past year or so where product discovery was important to me, where I was genuinely surprised that the solution was that easy and available. Heck, one was such a big benefit, I'll share: I changed the faucet hardware in my bathroom vanity, but instead of laying on my back and trying to turn the nuts by hand or squeeze in a pair of channel-locks between the sink and the wall, this time I got a targeted ad in my Amazon search results for the faucet advertising a basin wrench, which I didn't know was a thing. That's genuine product discovery, it saved me a little discomfort and a little time, and cost $6. I like the basin wrench, but if targeted ads didn't exist and I could pay a third party to provide me with the same advice I get from targeted ads, I'd value the service at maybe $2/year.
Targeted ads, especially as they're currently implemented, do have a cost to you as a consumer. It's largely invisible to non-technical users, but they do have a significant cost to your privacy. They cost computing cycles and battery life. They cost control of your devices, of your software, and of the websites you browse. Personally, those are pretty significant costs to me. And worst of all, IMO, they have a huge cost in mental fatigue, clamoring for my attention, because for every targeted ad that matches the "has this problem/wants to solve it with this product" combination there are 10,000 that fail one or the other condition. I value these costs pretty high, I bought an SBC just to run PiHole, I pay for Netflix so my TV is ad-free, I buy songs on my phone and wear noise-cancelling headphones in the shop at work so I don't have to listen to the incessant f*ing radio ads. Ugh. And there's little chance I can do anything about the billboards on my commute, though those are contextually targeted to my area rather than invasively targeted. I've probably spent hundreds of dollars avoiding ads, if I could pay someone hundreds more to make the entire industry disappear from my perception I would do so in a heartbeat.
That said, it's not possible to put a number on either the cost or the benefit that's true for everyone. Personally, I feel that those advocating targeted ads generally underestimate how much data is being collected, overestimate the benefits of product discovery, and underestimate the benefits available from alternative, less consumer-costly mechanisms for targeting, like content/context-based targeting.
How many dollars per year do you think the benefits of product discovery, as provided by non-GPDR, non-CCPA, no-holds-barred targeted advertising is worth to you? What price would you put on its costs?