Once upon an algorithm: storytelling in the age of AIOs

Not long ago, in a not very distant land, there lived a brand marketer with all the customer data in the world. Life was good. They knew what people were searching for, where they were doing it and how they could connect with them using their magic stories. 

However, now we’re through the looking glass. Even with the same data and the same messaging it’s become much harder to gain the same visibility and levels of engagement. It’s not that the audience isn’t there, but the ways in which users find information has changed. 

Previously, chapter one was generally a search engine and, over time, the results page evolved to offer multiple, branching opportunities for discovery - regardless, the path was linear and relatively easy to follow. Now, growing numbers of people go straight to an LLM and even when they use traditional search, few get beyond the AIO. 

That doesn’t mean storytelling as a concept is inherently under threat: the environment in which we tell stories has changed though. Consequently, we need to change where and how we tell them to improve the chances of our brand getting discovered and featured. We also need to think about the customer’s needs, LLM usage and engagement. Are brands creating content that matches the conversation?

That means catering to the needs of the AI crawlers as another audience. 

Skip to the end…

If your narrative isn’t clear, your stories won’t get picked up. But clear doesn’t have to mean boring; rather, it needs to be easy to interpret for everyone reading it, or from a bot’s perspective, scanning it. In practice, this will mean structuring stories around intent by anticipating and answering tightly defined questions.

While this means making the copy transparent enough for a non-human actor to interpret human motivations, it doesn’t mean we can forget we’re still writing for actual people too. So, we don’t need to dumb down the content and we absolutely should not be asking generative AI to write our copy.

Rather, we need to think carefully about an article’s structure and clearly signpost where an AI can find the relevant information. From this perspective, there’s a crossover between how AI reads and how most people engage with online copy, at least when they’re actively searching for information. 

Eye tracking and dwell time tell us that very few people read a full article when they are looking for something specific. More typically, they want to find the relevant parts as quickly as possible, as does the AI. 

Therefore, key elements to focus on are a headline that clearly outlines what the article will cover, subheads that highlight where the key information is located and a narrative that is structured logically towards a conclusion that clearly relates back to the headline. 

GEO is all about clarity and optimising for sense. In other words, there’s nothing to infuriate copywriters who have previously had to sacrifice their creative integrity by shoehorning in keywords at the altar of SEO. 

Forget the novella, think about the franchise

So, the rise of the robots doesn’t need to compromise how we write, but it does impact how much we need to write. A landing page, no matter how tightly written, isn’t sufficient to demonstrate your credibility on a given subject. AIs won’t take a single source at face value - they compare what’s said across multiple touchpoints to reduce ambiguity before surfacing an answer.

Visibility is holistic. That means every listing, blog, and social post - despite being published on disparate channels - functions as components of a wider whole. Consistency is key because the scrapers need to recognise that each piece of copy is demonstrably from the same source, which in this context is a good thing. Content, look and feel all have a part to play in making each element more than the sum of its parts. 

Where to start?

Understandably, this may sound alarming for marketers with limited resources. After all, we can’t possibly cover everything in this level of detail. The obvious place to start is to focus on content that will proactively support an active campaign or hero products, or you can follow the money.

By monitoring reviews and relevant forums on platforms such as Reddit, it’s possible to better understand what issues are of most concern or product lines that need better positioning or repositioning. In this way, it’s possible to develop hyper-relevant content that will address the most pertinent and timely search queries. 

One of our healthcare clients has even integrated AI to analyse inbound call tracking to identify what questions are most pressing to its high-value customers and pivoted its GEO content strategy to improve visibility in AIOs.

Look beyond the clicks

In an AIO-first world, chasing clickthroughs is no longer the priority it once was. This may mean having to educate colleagues when website visits inevitably drop off. Search is no longer a traffic source, it’s not a matter of ranking - but one of representation. Content is now about creating trust signals to give your brand the best chance of featuring in AI overviews.

Thankfully, we don’t need to rip up our big book of storytelling; this is very much an evolution rather than a revolution and your copywriters may well be (secretly) celebrating the demise of keywords. Actually, we need to go back to the basics of storytelling, narrative structure, and think more clearly about user intent - which we can easily monitor and measure through third and first-party channels.  

Ultimately, we’re always going to have to tell great stories regardless of whether our audience is human or not. That can only be a happy ending.