Instagram is experimenting. Here’s a short video recap of this post, if you don’t have much time to read. The platform is hiding like counts globally in tests to serve the community in a positive way. No more stress about likes. No more competition. How does removing this one button change the architecture of a platform? How does it change our interaction? Erasing a button impacts what we call “affordances” in platform studies – the offers and possibilities that a technology provides. So far, the results are positive, as the company stated on Twitter. But what are the potential drawbacks? And why do we love likes so much anyway?
Likes power platforms
Likes are intimately connected to web 2.0 platforms, which are generally characterized by friend/followers, favorites/likes, and profile options. These options have been around since the early days of Hives, LiveJournal and other platforms that set the tone for this architecture. Hiding likes means that the structure of social media as we know them changes. Their whole flow is based on keeping us on the platform, and making us consumer more. Likes are the backbone of what goes trending on many platforms, and what is deemed relevant, and what is filtered out. In the age of algorithms, cracking likes and follower counts is essential for all content creators, big and small.
Likes signify more than a number. They signify affect – emotion, hope, interaction, fear, hate. They should not be taken at face value. They are increasingly used in the data-driven strategies of platforms, and stand for our engagement. The whole architecture of a platform like Instagram or Facebook is build around this kind of engagement (including comments). Likes are oil. They keep platforms going, tell us which content is worth time and money. Likes feed the algorithms of YouTube and other platforms. If you know how to massage the algorithm, you go trending faster.
Likes are social proof
Numbers also have a social function. They tell us what matters in a community, and what doesn’t. In Hello World‘s chapter on data-driven art, Hannah Fry writes about stats and their social function elaborately. She discusses a study by Matthew Salganik, Peter Dodds and Duncan Watts (2006) who created a music player that had two versions – one with, and one without the download statistics. The scholars wanted to measure “true” popularity, but found that to be difficult. Between the hit songs and the absolute fails there was a gray zone of “success”.
The results were intriguing. All the worlds agreed that some songs were clear duds. Other songs were stand-out winners: they ended up being popular in every world, even the one where visitors couldn’t see the number of downloads. But in between sure-fire hits and absolute bombs, the artists could experience pretty much any level of success.
It’s in that gray zone of “any level of success” that certain songs went trending, but only in the version with statistics. A number of downloads, then, is a version of social proof. Based on their experiment, the scholars write, that social influence “contributes both to inequality and unpredictability in cultural markets” (p. 855). Numbers matter, in any shape and form, as likes, download stats and follower counts. Likes are a popularity tool that signify the opinion of others. But we do believe in others. If one of my friends liked it, then it is indicative of something. It must be good, interesting, relevant. There are different ways of doing social proof. Twitter for instance increasingly shows users the content that their friends have interacted with (with a small disclaimer which friend retweeted or liked it), but this content is not always produced by anyone they follow or have engaged with.
Ideally, likes go trending. Something that’s popular on a platform quickly can go super-popular or trending. Why? Because it has what media scholar David Nieborg calls this the network effect, or the winner-takes-all effect. If you hit a certain stage of popularity, you’ll show up more in recommendations, news feeds and walls, you’ll get even more likes, and go viral. But as any influencer knows, cracking that network effect is not easy.
Likes predict
Likes have a predictive function, especially now that platforms increasingly automate and filter content. In this sense, algorithms and likes are also easy and comfortable – they lead to more of the same content, and can create happy filter bubbles. And we don’t think twice about it because these options are deeply ingrained in platforms as we know them today. I trust YouTube to some extent. If the platform sorted this ASMR video out for me, then I often give it a go before going to bed. And it’s not often far off, because to some extent I trained it.
Likes rank content
Instagram puts mental health first, and it’s an interesting business decision to experiment with the like button more. The hot take that I see in a lot of journalism around Instagram’s tests is that social media put a lot of unwanted stress and competition on their users. Removing like buttons will improve mental health, PBS for instance writes. This will cure addictive behavior and re-discipline users. While such claims are extreme, I do think that removing likes creates a different type of community.
Likes rank people and content. They are a value claim. Popularity, in other words, seems to stand for quality. The most popular tweet in terms of likes and retweets must be a good tweet, a valuable tweet.
But likes are murky and fickle. Why does this one Facebook photo gather 40 likes and the other one 200? For consumers and professionals, it’s sometimes hard to judge why something is successful. It gets under our skin. Likes create a distinct kind of trivial paranoia. Why does a similar video of lower quality get so many comments and likes? Why does this other photograph on Instagram of Utrecht go trending, and not mine? This is what creates the stress for many users, and maybe removing the button will indeed have some positive effects.
What’s next?
Social media are hierarchical and competitive partly because of these rankings. They are also much like a game – we love seeing those numbers go up. it gives instant satisfaction. It’s not that different from a game of Candy Crush, really.
Instagram is at least reflecting actively on its architecture, and working towards innovation. We can do better. Likes are not the only way to measure engagement or to identify content. We can navigate through hashtags, comments, and followers without ranking ideas and people. Engagement only matters from a business perspective, and it’s not the whole stories. Likes are part of engagement in a sense that they can tell us which content can be monetized, and which cannot. But dealing with digital content is much more complex. It can’t be watered down like this.
If we want to improve platforms, we need to move beyond the typical web 2.0 features which have become, frankly, dogmatic by now. I’m curious to see what these experiments will lead to. Sign me up, Instagram.
Nicolle Lamerichs 