[PaulHoule]

Tracking is necessary for the advertising economy to control bad behavior on the part of publishers and advertisers, not just to serve targeted ads.

No matter what there would be discrepancies in the numbers (publisher says it sent 75 clicks, advertiser says it got 70) and that breeds mistrust. Participants have a reason to lie. Having multiple third party watch the whole thing helps them trust each other.

[masswerk]

This has been possible before. Google was built around content based advertising. Also, maybe click-through rates are not that important? Maybe exposure is a more decisive metric? As a side-effect, we may reduce social bubble effects and maybe even return to a shared reality? (Many of the unwanted effects of the Web are really due to targeted advertising and its consequences.)

[aerosmile]

You're clearly under the impression that all that Facebook does is get you good click-through rates and low conversion rates. Similarly, you're under the impression that Google's conversion rates are through the roof. Did you ever stop to ask yourself if your world view might be limited to a small sample size (eg: your own experiences)? I know plenty of businesses where the conversion rates are exactly the opposite of what you described. It turns out, the right marketing platform is related to the product you're selling. For example, if you want to buy a vacuum, Google will do a great job in connecting you with the right advertiser. But what if you just started a new hobby and don't yet know what it is that you need in that hobby? Eg, you started racing, but have no idea that upgrading your suspension will get you more performance than upgrading your exhaust? The advertiser selling suspension components will use Facebook to share this information with you (and Facebook will do a much better job than Google of identifying the person who needs that component but doesn't know it yet), and they will also advertise on Google as well, but Google will only be relevant once the buyer starts researching suspension solutions. Kind of nice to be the first one to pitch a suspension solution to a willing buyer, don't you think?

[masswerk]

On the other hand, if I'm not on FB, I probably miss those vendors of suspension components all together. Advertising "in the bubble", as opposed to "in the world", comes with its disadvantages. With content based advertising (media analysis), you'll probably catch me at the related watering holes and communities. Also, excluding non-targeted audiences doesn't exactly benefit a shared world view.

P.S.: The general idea of targeted advertising misses the concepts of state of mind and focus of interest entirely. There's a (significant) difference in delivering a message in context and out of context. (The latter may have even adversarial effects.)

[aerosmile]

> Advertising "in the bubble", as opposed to "in the world"

Is this an argument for the open web, or is this an argument for Google being more popular than Facebook? If the former, I am with you. If the latter, the gap is not as big as you might think - 4 billion worldwide users of Google Search vs 2.85 for Facebook [1]. Slight advantage to Google, but how many US advertisers really care about international buyers? When it comes to advertising, you want people with money burning holes in their pockets. Facebook's financial results show that they plenty of access to this demographic.

> The general idea of targeted advertising misses the concepts of state of mind and focus of interest entirely. There's a (significant) difference in delivering a message in context and out of context.

You're slightly lagging with this argument. This exact reason was why Google was so dismissive of paid social back in the formative years of Facebook ("who gives a shit about what college students talk about on social networks?"). And then the oh shit moment happened in 2011. Eric Schmidt had to step down and Larry Page tied bonus payouts to the success of Google Plus. You have to give Larry some credit here - while a bit slower than Zuck, he did see the writing on the wall before the rest of the world did. If you tried advertising on both platform in 2011, Google would just crush you with their results. By 2016, Facebook was already competitive with their interest and lookalike targeting for a large group of advertisers (and even enabled Trump to win the presidency), and by 2018 the game had advanced to the new concept of "creative targeting" which is basically entirely driven by the algorithm and takes minimal targeting input from advertisers. At that point, paid social became so good that it got creepy and turned the public sentiment to negative and incentivized Apple to jump on the band wagon with iOS 14.

So I wouldn't agree with your argument that Facebook has a fundamental problem with recognizing intent. If anything, they are too good at it for their own sake.

[1] https://review42.com/resources/google-statistics-and-facts/

[masswerk]

I wouldn't pose this as FB vs Google. Google is dealing intrinsic conversion metrics, as well, which are highly problematic regarding the total impact. (So, Google is yet another, while maybe broader bubble.)

On a historical note, as it turned out G+ was soon scrapped and quite a loss, while we're dealing still with the fallout of Google's uniform platform strategy, which was put in place to provide for G+.

[PaulHoule]

Some kind of verification is necessary for impression-based advertising too.

For instance, there are discontents around Nielsen (they've had scandals in India, and I infuriate people in the TV industry with the suggestion that a Nielsen home got bribed to blast MTV in an empty room) but the participants believe in Nielsen: people know probability-based sampling basically works.