The biggest shift in social media marketing over the past five years hasn’t been short-form video, or even the development of generative AI. The key evolution, from a marketing-specific perspective, has been the move away from interaction and towards straight entertainment.

Now, you could, of course, argue that entertainment has always been a key element of marketing. But in the early days of social media, a central focus for brands was sparking interaction and driving engagement, by providing a means for your target customers to get in touch and get responses from your brand presence.

The implementation of algorithms made this even more of a focus, because more comments and likes resulted in more reach. And while engagement is still a factor, the platforms themselves are now moving beyond prompting likes and comments, and are refocusing, instead, on time spent in each app.

Which is all because of TikTok.

For years, Meta held the reigns in social media, and as such, it dictated the trends and norms that guided how users interact. Meta had learned that comments and interactions were what prompted users to stick around, so its algorithm prioritized posts that garnered more of these behaviors.

Which inadvertently incentivized more divisive, argumentative takes, because the content that’s most likely to spark response is that which drives high emotional resonance. Joy, fear and anger are the key emotions that prompt response, and unsurprisingly, the algorithmic shift led to these becoming the main drivers of online activity.

But TikTok flipped the paradigm, and has changed the way that all platforms now operate.

Rather than relying on your connections, TikTok emphasized public posting, and rather than focusing on likes and comments, TikTok focused on time spent, and taking in more indicators from each clip to then refine the recommendations that it shows each person.

Essentially, TikTok re-aligned the social media matrix around entertainment instead. Which is huge, because the actual ‘social’ aspects that had long been the focus have now been relegated to a secondary plane.

And that also means that social media marketing strategies need to be re-aligned in step.

Don’t believe me? Need more convincing?

How about the fact that 50% of the content that users see on Instagram now comes via AI recommendations?

These are not posts from your friends and family, and this is not content from accounts that you’ve chosen to follow. Also, Instagram isn’t as concerned about whether they’ll drive likes and comments, the key measurement is how long you spend watching, and how fast you skip by each (or not).

Facebook usage is also up, and 30% of the content that users see on Facebook also now comes via the same AI recommendations system.

But the key consideration here is this: While time spent on Facebook has increased as a result of these topical recommendations, creation and engagement continues to decline, with fewer people posting to both Facebook and Instagram than they have in the past.

That’s particularly true among younger audiences, while notably, usage of Instagram Stories is also in decline, down more than 10% on previous levels.

People are spending more time consuming AI-recommended video clips, but they’re increasingly less likely to comment and engage beyond viewing. As such, social media is becoming less and less social, and is more like traditional media, with interactive elements.

The consideration from a marketing perspective is that you may need to shift your focus in line with this, by shifting away from posts that spark comments, and more towards pure entertainment as a result.

Triggering engagement has long been a key tenet of social media marketing, because as noted, platform algorithms have prioritized content that generates direct response. But now, virtually every platform is increasingly aligning with time spent, which comes, mostly, via video clips.

And in that scenario, the most entertaining content wins, not posts that prod users for their thoughts.

And when you also consider the rise of generative AI, that’s likely to become even more of a focus, because while generative AI tools can come up with human-like images, text and video clips, what they can’t come up with is truly creative concepts, or at least, it’s not as so easy to make them give you such, without having a concept in mind for yourself.

For example, I asked ChatGPT to give me an idea for a “wacky, funny campaign to promote a new lemonade brand.”

Okay, #SqueezetheDay is not horrible, but the tag itself has already been used by many juice bars and designers in both on and offline promotions. Which highlights the key issue in using AI generators for such purpose, in that most of the concepts that such systems will produce are going to end up being fairly generic, and therefore unlikely to resonate in the new, entertainment-focused social media environment, where concept, increasingly, will be key.

The challenge for AI generators in this sense is that they’re inherently derivative, as the system can only iterate on what’s come before, so everything that it spits out is going to feel somewhat cliché and repetitive.

Because it is, the system is just giving you a variation of what it conceptually understands from the terms that you enter. It’s going on “lemonade” and “promotion” and pumping out concepts based on what it can match up, based on previous uses of these terms within a certain proximity.

As such, everything that comes out of an LLM-based system will likely feel a bit stale, whereas the content that’s most entertaining is fresh, interesting, and generally a new take on past ideas.

Is it possible than a generative AI system could produce a great idea? Sure, but systematically, that’s not really how they’re designed, which reduces the likelihood that an AI tool will be the source of the next great concept.

But regardless, AI tools are being used more and more. Which could actually further enhance the value of truly unique, interesting concepts, amongst a rising tide of past ideas that have been regurgitated by AI systems.

Essentially, entertainment, and increasing time spent with your content, is what’s going to drive more reach in social apps, as the platforms shift further away from their social roots, in line with user behaviors.

Past best practices, including engagement bait, hashtags, calls to action, all of these are going to become less important, as more platforms shift to focus on AI recommendations.