As far as I understand how things like facebook or reddit work they:
-
offer an unpaid service to mass consumer
-
harvest data of the people who use the service
-
offer paid advertisement space to companies
-
companies buy advertising because the vast data promise precise targeting
-
precisely targeted ads convert into sales for companies
-
the ROI (profit gained to cost of ads) when buying social media ads is greater than ROI on tv or whatever other ads
-
social media expand on the profits gained from ad space sold to companies
-
social media corp announces a brand new feature and we return to point 1)
Which step is the closest to breaking? Where are limits of growth and who hits them first? Is there a cap on marketing budgets beyond which companies won’t afford social media ads and tech corps won’t afford expansion and maintenance? A cap on how much data (=how precise ads) can they harvest from us? A lower threshold of general wealth below which ads won’t convert into more profit because people are too poor? A breaking point of enshittification at which user count (=ad visibility) plummets?
The recent apeshit of tech companies after the raised interest rates made me feel that the entire thing is quite fragile and ripe for falling… But I’m not a financial advice so maybe I’m completely clueless.
Step 4 is most likely from a markets perspective and step 2 from a regulatory perspective.
The regulatory landscape is slowly shifting against limitless surveillance as the cost of using the free internet. Free platforms like Reddit and Facebook will struggle if they can’t collect limitlessly. More interestingly for me, portability and interoperability regulations could lead to exoduses from those platforms as people are no longer locked in - there is no reason why you can’t message someone on Signal using Messenger other than Facebook doesn’t want you to. Taking your personalization data from Spotify to Apple Music would mean you didn’t have to start from scratch if you didn’t want to keep paying Spotify. It’s an existential threat to companies that depend on a “data moat.”
From a markets perspective programmatic advertising is (as Cory Doctorow says) a bezzle. It’s a scam. ROI and attribution rely on faithful and accurate data on audiences and reach, in addition to engagement. The major ad platforms have regularly been caught lying about these. To be clear paying for ads on these platforms largely works - that’s not up for dispute and is why the bezzle is successful. The issue is that these platforms are selling targeted ads (I.e. show ads to people we believe want a pair of Nikes) that work BETTER than contextual ads (I.e. show ads to people searching for shoes) and that’s not clear.
I think messaging is more likely to see progress made on this front. I don’t see regulations forcing Spotify to allow for their personalization mechanisms to be transferable and that doesn’t really make sense why they should have to, while messaging I can see happen even if only due to popular demand.
It’s a more nuanced argument to be sure. The premise is that your music preferences are traits and knowledge about you that Spotify learns when you use their platform, just like how Google learns you like chips and Facebook learns you went to school with Mike. It’s not really fundamentally different to learn about music preferences than it is to learn about shoe preferences, it’s only different in how they’re used.
When you switch mobile carriers there is a level of interoperability baked in. You keep your number, you can transfer phone contacts from one service to another, etc. Tech companies have everyone convinced that their old tricks are so clever and new that they need new rules, is all.