From friends to feeds: How social media got weird

Opinion

The tools to advertise on social media have never been more powerful. And yet, they risk adding to the ever-growing volume of work that looks exactly like everything else around it.  What can advertisers do, asks Chris Herbert-Lo?

Viagra was created to treat high blood pressure before certain side effects became evident. Play-Doh was made to clean wallpaper before pivoting into a children’s plaything. And social media started out as a way to connect with mates before becoming the algorithmic content-delivery platforms we know today. 

When it first exploded in the mid-noughties, the social media user experience looked very different. Photos from parties, uploaded in batches, friends meticulously tagged. Profiles curated with a song that auto-played whether you liked it or not. And whatever the hell was going down on Bebo.   

The advertiser experience looked different as well. From paying a small fortune (in early digital terms) for a branded Myspace page, to right-hand column ads on Facebook that were little more than a badge with a logo on. 

Two decades on, the advertiser experience has improved beyond recognition. I’m not sure the same can be said about the user experience. 

An onward march for advertisers

Part of the well-understood story of media fragmentation is that platforms such as Meta, X and TikTok have grown to fill any moment of downtime. There was once a very real debate about whether social platforms would successfully transition to mobile. In the end, they didn’t just transition to smartphones; they conquered them. 

For advertisers, this means vast audiences, endless formats, and granular targeting, all available at the tap of a screen. And while there is concern about fleeting attention levels, recent research suggests repeated short exposures are a perfectly good substitute for the lack of attention to any one ad.

In other words, don’t worry if no one’s really watching your message in full; they’ll absorb it eventually. 

All of which perhaps explains the confidence of Mark Zuckerberg, who recently outlined a vision of advertising so frictionless it barely requires human involvement at all:

We’re going to get to a point where you’re a business, you come to us, you tell us what your objective is, you connect to your bank account, you don’t need any creative, you don’t need any targeting demographic, you don’t need any measurement, except to be able to read the results that we spit out.”

When the algorithm becomes apparent

For users, the story has been slightly less utopian. Platform growth moved from new users to keeping people scrolling for longer. Out went actual friends and in came algorithmically curated feeds, serving up content based on what you, or someone like you, engaged with previously.

In terms of how that’s going, in just the last couple of weeks, we’ve seen:

* Landmark cases in LA and New Mexico, which found social media companies liable for harm caused to children due to their addictive nature. Opening the floodgates to hundreds of similar cases. 

* A BBC investigation in which whistleblowers allege social media algorithms allowed ‘borderline’ content (e.g., abuse, hate speech, sexualised material) to spread because it increased user engagement. 

* ‘Content creators’ of the Manosphere showing Louis Theroux just how well they’ve figured out what works for those algorithms. Spoiler alert, much of it seems to be ‘borderline’. In the words of one creator, “If I were a good person, I wouldn’t have blown up on social media”.  

Whilst tighter age controls now look likely, no doubt leaving social platforms scrambling to develop their version of a Lost Mary, there’s a question of how much this all affects most users.

For many, none of it will be especially shocking. Much of it now feels baked into the experience. 

But as with the fable of Al Capone being taken down by the taxman, the social platforms may face a problem far less headline-worthy. Boredom. 

Mintel’s ‘Anti-Algorithm’ trend for 2026 points to a growing fatigue with feeds that feel like endless variations on the same idea. Its research found that 60% of UK users who follow influencers believe they often recycle other people’s content. Which perhaps points to an algorithm working exactly as intended, and people being turned off by it.

How does advertising react?

For advertisers, this presents a slightly awkward paradox. On the one hand, the tools have never been more powerful. On the other hand, they risk adding to the ever-growing volume of work that looks exactly like everything else around it. 

Zuckerberg’s vision of fully automated advertising sounds a lot less appealing if it’s simply generating more of the same stuff to populate already formulaic feeds.

Of course, some of the algorithm’s rules must be followed. Mass market approaches trying to gain appeal across a broad group of people won’t cut it in a world built on niches and micro-interests.

But we are an industry built on difference, one that loves to highlight a ‘sea of same’ and break category conventions. Which surely means resisting the algorithm’s gravitational pull just enough to make something that doesn’t feel entirely pre-approved by it.

If the machines are going to do all the thinking, the least we can do is stop helping them to be boring. 

Chris Herbert-Lo is a strategy partner at the7stars. Read his new monthly column for The Media Leader on the first Tuesday of each month.