I work in adtech, and built a product tried to measure the impact of advertising, that Google/Facebook/Twitter all immediately copied. So going to answer from m

I work in adtech, and built a product tried to measure the impact of advertising, that Google/Facebook/Twitter all immediately copied. So going to answer from my experience. > I could never figure out if advertisers don't know what's going on This is largely not true. The big advertisers all have a pretty good idea of which spends were accomplishing their goals better than others. It's the goals that might not be obvious from the outside; everyone always assumes click-through or call-to-actions but for the big spenders that tends to be pretty low on the priorities list. > , or they merely don't want to know. This is also true. So while everyone knows what the underperforming ad products are, no advertising executive wants to be the one sticking their neck out doing the work proving advertising _doesn't_ work. That's their livelihood, once you question the necessity of an established market leader, you're more likely to invite question the necessity of you instead. Is it "Google advertising doesn't work" or "you don't know how to make Google advertising work"? > I get the feeling that companies value brand perception so heavily they don't know they're getting ripped off by the distribution networks. When we asked major advertisers and agencies what their top priorities were: all of them put brand safety at the top of the list. Above reach, above cost efficiency, above click through, above conversions. Brand perception is a subcomponent of brand safety. It takes years to build up good brand perception and very few mistakes to undo that progress. > Or, individual ad executives just want people to talk about their ad at the water cooler, so don't purchase ads to most efficiently use their employer's marketing budget, but rather maximize impressions to inflate their personal brand. Most ad executives never hear much about their ad campaigns. But if they do at the water cooler in a positive light, you've probably made their whole career. > Like I said, I don't know; maybe everyone knows about all of these things and they think untracked ad impressions are worth what they're paying. It's a mystery to me, but I'm somewhat surprised the traditional ad industry is alive and well. Tracked ad impressions are relatively a very new thing for this very old industry. Generally at the scale advertising works at, they don't need full comprehensive impression tracking to know if they're accomplishing their generally unambitious goals, it's easy enough to get a representative sample and extrapolate with statistics. If you're trying to sell something, it's way easier just tracking how many sales you got than trying to divine how impressions and click through rates could impact sales. There's a bunch of A/B testing going on with how campaign spends are broken out. Different products or regions can be intentionally left unadvertised to use as control to measure effectiveness. Basically there's a lot of data (and very few it from the adtech platforms themselves, since they had to know this before the era of digital advertising) that supports advertising as a whole works really damn well, and most of the work is just determining what advertising strategies are working better than others. Plus, the data from Google/Facebook is largely junk anyway (same as everyone else to be fair). Just check out your advertising profile on those respective sites. You'll probably get a big laugh on the accuracy of your interests.