I find it unbelievable that Facebook doesn't have the fake profile situation under control. Facebook builds an incredibly detailed social graph of every user (and non-user) with a big trail of activity, on and off Facebook.
Surely there are signs; surely there are common characteristics, and if this journalist can write such a detailed exposé with only public data, Facebook can do much better.
Twitter has a similar problem with bots. They're bad for the platform, but the short term incentives are all wrong for them to actually do something about it.
I'd think because at least publicly they've placed a great deal on emphasis on accounts being real, with the stated goal of producing a certain kind of environment. If your social network becomes polluted with trash and even brutal scams, then won't engagement go down? But I guess they've run the numbers.
Because as it becomes more fake they lose their reason to have users in the first place. Ads are the product, metrics are the data that leads to the product, and users are the raw material. If they lose the raw material the product loses value.
Yeah, but surely they must understand (especially after their uncomfortable turn in front of a congressional committee) that if they keep up the bullshit for too long they are going to find themselves regulated.
There may be no immediate financial incentive, but I'll bet good money that they fear a future of regulatory compliance.
I doubt it. User growth is used as a metric to measure future ad revenue. Fake profiles aren't going to generate a lot of revenue from those ads. Facebook would simply need to make a statement about deleting a bunch of fake profiles, but they don't expect revenue to dip.
However, that is not the case for a pre-revenue or pre-profitable business. Those startups are dependent on VC funding, they might lie about users and usage to get more funding (see Silicon Valley Season 3). *This is 100% inappropriate and should not be done. You will burn credibility with investors, for your current venture and any future ventures. It's better to shut down a venture and move on, than to lie about your metrics.
This morning a guy from high school finally joined Facebook and sent me a friend request. I don't track him in real life, but were have enough of a connection that I'll click yes if that is really him. I honestly do not know for sure if this is him or not: I decided it is, but it wouldn't be hard to social engineer all the data that is on Facebook so I'm not sure. (I decided it is him because his name isn't in my friends list, while all the spam I've seen so far is someone already in my friend list)
To have this many accounts, they might have a series of scripts and potentially farm out captcha. I find it interesting the article mentions one of them messing up and switching to a fake account in a post. That would suggest some of this is done manually, which would be bizarre.
I'm sure Facebook creates filters that helps delete some, but remember it's a constant war between spam script writers and platforms. They're constantly finding ways to get around each other.
Surely there are signs; surely there are common characteristics, and if this journalist can write such a detailed exposé with only public data, Facebook can do much better.