I’m no fan of Meta and its monetization of user data. That said, there are some things worth noting about Threads, Meta’s new social network/Twitter clone.
One of the challenges of starting services like Uber, Twitter, and others that rely on network effects is solving the cold start problem. Network effects mean that people get more value as more people use the network. The cold start problem is this: when a network is first launched, there are no users. If there aren’t enough users on a social network and no one to interact with, everyone will leave.
As Andrew Chen explains in his book The Cold Start Problem, successful network effects require both the service and the people using the service.
For the service, Meta runs Facebook, Instagram, and WhatsApp. Creating a service that can accommodate millions of users on day one holds no secret for them. They also have the infrastructure to do so. For users, Threads solved the problem by allowing users to import their Instagram connections with zero friction, so they could find value in the service in no time. Threads grew to 100 million users in just a few days, which is insanely fast.
How is Meta monetizing Threads? For now, Threads has no ads or uses any of Facebook’s famous addiction-inducing devices. But ads and addiction-inducing strategies will come for sure—hey, this is Mark Zuckerberg running the show. Make no mistake, Meta is the best at ad targeting and exploiting its users, and Threads will not be the exception to Meta’s portfolio.
Meta has announced that Threads will support ActivityPub, the open social network protocol used by Mastodon. It’s not clear how ActivityPub will benefit Threads users, considering that the total number of users using a network that supports ActivityPub is relatively small. Consider, for example, that as of March 2023, Mastodon’s fediverse has only 10 million users1.
The ability to jumpstart a new social network by leveraging the social graphs of their existing services, together with the ability to spin out services that scale massively are serious advantages for Meta. They not only allow Meta to create new services that allow additional monetization via advertising but also that generate data for training their AI models without depending on Twitter’s or Reddit’s data. OpenAI may have trained their models using data social networks for free but that won’t happen again.
cfr Statista, Number of registered Mastodon users worldwide as of March 2023. ↩︎