People like getting services at no direct cost, advertisers like selling to those people. While most want services at no cost to them, they don’t like being tracked. Modern ad tech puts people into a kind of human terrarium. Large tech platforms from Alphabet (once Google) and Meta (once Facebook) down the line gather and cross connect absurd levels of data about consumers.
When I write about advertising, I cite as my credential that I’m a demon at an ad agency. When I am quoted in media, like Texas Observer, I am a managing director of Push ROI. Most people will not view those titles very differently.
For most tech platforms advertising is both Samson’s hair and Achilles’ heel. The source of strength and the single largest vulnerability. All that data collection, benefits the platforms, but no one else. I’ll explain more but first…
I’m trying a new approach to content syndication.
I typically write and publish 1-4 weekly articles. I once syndicated the full text of these articles on large online platforms where people already spent time. In concept so doing maximized distribution while maintaining control over the work.
However, as large platforms continue eating the web, it is clear unfettered broad syndication of everything I publish is just feeding a machine I don’t like.
This article was written for broad syndication, along with links to my other recent work. I’m also linking to some interesting things I’ve been reading as I believe doing so makes the internet less platform dependent.
This article and read by AI podcast is a eulogy to Google+, and a lament about how on social media brands want to be people, while totally normal people stress out over their brand image.
People strive to be brands creating a loss of self, and often a loss of any potential brand. Brands try to be people, in programmatic and contrived ways. Both are organic as a Twinkie asking how the fellow kids are doing.
For Push ROI I wrote that NPR as an organization ceasing its use of Twitter will have negligible impact on NPR.org’s monthly traffic. The site seemingly receives about 1.68% of it’s traffic from Twitter, and NPR’s owned Twitter accounts likely drive under 0.20% of NPR.org’s monthly traffic.
Not in the article: Twitter is losing a lot. NPR podcasts collectively garner over 20 million monthly listeners, and in broadcast NPR stations reach 23.5M listeners each week. The value of many of those shows simply saying “follow us on Twitter” almost assuredly drove more traffic to Twitter than Twitter ever sent NPR.
Back To The Story
We’re on the precipice of 80% of digital advertising spend routing through four companies, Alphabet, Meta, ByteDance (TikTok), and Amazon. These companies have a lot of first party data, and a lot of information from data brokers. But these ads aren’t better for media companies.
In an empirical study on the impacts of behaviorally targeted advertising on online publishers’ revenue, researchers at the University of Minnesota, University of California, Irvine, and Carnegie Mellon University found that publishers only earn about 4% more revenue for ad impressions that had cookies enabled than for ones that don’t. The study is from 2019, so cookies are mentioned a lot. But I have seen no evidence that the conclusions were false in 2019, or that anything has changed as of 2023.
The targeted ads aren’t really better for advertisers, at least not like they once were. Next month I’m giving a talk about advertising to a group of small business owners. I was asked if I could speak about Facebook ads for small business. A topic I frequently offered trainings for from late 2012 until 2015. I’m not speaking about Facebook ads, specifically because the ads platform got nerfed.
Facebook — in part because people like me talked about how much data could be found on the ads platform — has hidden a lot of options. The company has more data than ever, but advertisers cannot directly access and control that targeting. This keeps Facebook from being scrutinized by watchdogs, it also helps Facebook earn more.
Advertising via Facebook once allowed incredibly granular control over ads targeting. Now ads on Meta is simplified, and machine learning attempts to improve ads performance obfuscating much of the process. The ads are also more expensive now.
In 2022, U.S. District Judge James Donato ruled that a lawsuit first filed in 2018 accusing Meta of deceiving advertisers about “potential reach” of ads could proceed as a class action. Having seen Facebook inflate audience sizes since 2013 I fully believe Meta did so, and did so deliberately. If an advertiser wants to target 100 people, but Meta charges them to show the ad to 200, Meta earns more.
This is something Cory Doctorow calls enshittification. A pattern of platforms he describes like this,
Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die.
For ads platforms the more simple the UI, and less transparent they can make the process of ad buying the more money they make. That’s why I called it Samson’s hair and Achilles’ heel. Consumers gave up so much data, all so advertisers could pay extra for microtargeted highly relevant ads, that are deliberately being served to people outside the targeting block.
TikTok is not a benevolent and wonderful actor, but the reason U.S. big tech wants them ban from the country is to protect existing monopolies. I’m not proposing the solution to every problem with big tech and ads surveillance, but it seems like neither is anyone in government. Breaking up monopolies, or creating comprehensive privacy legislation are solutions. U.S. leaders are asking about Wi-Fi, and Sci-Fi.
Like many I used social sites, mostly as a means of amplifying, and saving interesting articles, and videos. After my unofficial banishment from Twitter I added a section on my personal site where I keep bookmarks and notes for the same function. These are the direct links to a handful of the articles I took note of this week.
- What The Heck is Happening At Uber? (Alex Kantrowitz / Big Technology)
- Dril Is Everyone. More Specifically, He’s a Guy Named Paul (Nate Rogers / The Ringer)
- Italy’s new rules for ChatGPT could become a template for the rest of the EU (Ioanna Lykiardopoulou / The Next Web)
- Banning TikTok could weaken personal cybersecurity
(Robert Olson / The Conversation)
- Tesla sued over creepy spying voyeur employees (Louis Rossmann / YouTube)
Podcast (audio articles)
I’ve been adapting more of my articles as podcasts for accessibility and ease of consumption. These are available in most of the normal places.