Improving Twitter's Follow Mania
I’m fascinated by the follower/following dynamics of Twitter. There’s no formal protocol for any of the platform’s social norms, but stick around long enough, and you start to understand the informal ones. It’s good form to follow someone back if they follow you. Unfollowing someone can be perceived as a slight so serious that it would’ve been better to never have followed the person in the first place. Don’t follow too many people because following too many people with too few followers of your own is a bad look.
The rules change for so-called “influencers” who have ratios that demonstrate the power law distributions in audience capture that emerge inherently on social platforms. While most users have tiny follower counts in the 10s, 100s, or maybe 1000s, influencers can have as many as 5+ orders of magnitude more, with the follow counts of the likes of Barack Obama, Justin Bieber, and Katy Perry ranging in the 100s of millions. Unlike with their closer peers, Twitter users follow influencers not expecting to be followed back, and help amplify their signals to the outer reaches of social space.
Especially where Twitter is concerned, this system leads to profoundly unhealthy behavior. Leaving aside the platform’s more serious problem of sowing division and outrage, people obsess over their follower numbers, and make follow decisions not by the quality of the content users produce, but how it will look both in terms of their own follower count and who those users are. Normal users follow their friends, and the influencers who embody an ideological alignment that reflects their own, but only to the point where it doesn’t make their ratio too disproportionate. Influencers don’t follow anyone at all (or if they do, only other influencers) because it’s a bad look to follow plebs, and it might make the other plebs jealous.
Is there a better way? Yes. And some of the fixes are surprisingly simple.
I’m a relatively new entrant to Instagram, having used it for the first time ever about five months ago 1. On the surface, the following/follower system is basically the same, but there’s an important twist – it’s really hard to check whether someone follows you back. There are ways to do it, but the platform’s made a deliberate choice to obfuscate this information as much possible. Unlike Twitter, there’s no follows you badge under a person’s profile to make it easy.
This turns out to be a really good decision. Instead of users agonizing over how many followers they have and carefully following only the people who followed protocol by following them, they follow accounts that produce content they like. There are exceptions of course (notably new users will probably be carefully tracking every single follower they have), but in general people don’t worry about it as much. Once you following/follower counts are in the hundreds, it’s not worth the effort diving in the lists to micromanage unfollows. This leads to subtle benefits:
- There’s less risk that other users will take an unfollow personally because they won’t even notice. Users can unfollow more liberally.
- It also means that users can follow more liberally for the same reason – less chance of hurting someone’s feelings if you unfollow them again later.
- With more liberal following and unfollowing, users can take more risk in accounts they follow, including new users with small follow counts that have a hard time gaining traction.
- So all in all, users are more likely to be following accounts producing content they actually like, rather than people they don’t want to piss off.
None of these are hard rules, but the point is that generally healthier behavior is being gently encouraged.
Another interesting case is Reddit’s follower system, which most people probably don’t even know exists. It deviates wildly from Instagram and Twitter in that users can’t see how many followers any other users have, and the number of users an account is following is hidden completely, even from the user themselves. Here’s my profile pane, with a count visible only to me:
Of course Reddit’s follower system doesn’t do very much – basically, it causes an account’s self-posts to appear in the timelines of people following them – but even so, it may be the healthiest design of them all. It’s still possible to use the engagement on a self-post – the number of upvotes and comments it gets – as a proxy for how popular a user is, but the “high score” follower and ratio numbers just don’t exist, so there’s nothing for users to torture themselves with in pursuit of winning the grand popularity contest of the internet.
Lack of follow counts is doubly good because the ones found on Instagram and Twitter are often highly misleading. It’s not unusual to see someone who looks like an influencer with 10s of thousands of followers produce tweets that regularly get only a handful of likes – highly disproportionate with the engagement you’d expect given their ostensible popularity. This is usually because they were popular at some point in the past and are still followed by many now-inactive accounts that didn’t unfollow before succumbing to cold sleep, or more simply, they paid for an army of bots. I personally know people in both those situations 2. The amount of engagement a user gets on their content is a much better indicator of their popularity than their follower count is, just like Reddit.
Unfortunately, despite all this rumination, the truth is that platforms like Twitter have little incentive to fix the defects in their systems, even if it’d be healthier for their users. Desperate vying for more followers and more popularity keeps people online and engaged, and thanks to ad-driven revenue models, that translates to money in the bank.
1 Okay, technically the account has been open for a few years, but I only used it as a placeholder and had ~0 followers.
2 And actually, my Twitter account is kind of like this, minus the thousands of followers thing.
September 5, 2020 (4 years ago) by