Crafting Content: The Human Touch or AI Efficiency?

The surge in AI-written content is hard to ignore, but human expression is enduring, endearing, ineffable. Can the two harmoniously coexist in a marketing strategy? And if so, how?

We can celebrate (or blame) Benjamin Franklin for content marketing when he published an almanac to market his printing business and kicked off the whole thing. After that, content marketing was seen as a strategic way to distribute valuable, relevant and consistent messaging to attract and retain audiences.

These days, those of us in marketing try to keep up with the insatiable appetite for more of it. Like a bottomless bag of potato chips, consumable content is about feeding a voracious beast that’s never full – and never stops at just one bite.

Enter Generative AI, the digital prodigy that swipes away the heavy lifting of writing content. Not to mention the cost. A quick prompt and the copy vending machine will churn it out. No writer required.

Slight Digression

I’m a writer. I believe that anyone who puts themselves on the page, whether to sell soap or publish the next great Canadian novel, is doing something disciplined and artful. Because of that, I’m sensitive to the differences between merely serviceable and stunning writing.

When AI-generated copy is submitted to me, I feel the same as when my cat drops a dead mouse at my feet. I appreciate the spirit of the gesture. But ultimately, it’s lifeless and a bit yucky.

I get it. AI is the future, now. And it’s addictively easy. ChatGPT removes barriers so that anyone who struggles with writing feels like they can communicate and access better opportunities. My business mind also understands. Content is expensive and time consuming to create, and it’s hard to justify investing in it. AI is a game changer.

I wrote this article because I wanted to make peace with AI, assess its value, and use the tool without bias. First observation, that little AI had parents who didn’t love it. How else could they name their baby Generative Pre-trained Transformer (ChatGPT)? So I gave my AI a name. Karl (he/him).

The Either/Or Syllogism

Karl is fast and efficient, no argument. As a productivity tool, he’s going to radically impact the bottom line. Generative AI works by sweeping the entirety of the Internet to find patterns in data and then uses them to predict word combinations.

Because of that, what’s produced often sounds sort-of-ok-enough. But deep down, don’t we often come away with the ironic feeling that our robot writers lack the human touch? Humans have the emotional intelligence that Karl doesn’t have (…yet), something that’s often revered for having greater value.

This has become the crux of the debate. Choose Humans or Choose Karl. The rock-em-sock-em battle lines are drawn.

The Fact is, They Serve Different Needs

When it comes to AI versus human-generated content, both have pros and cons for their efficiency, quality, cost, and online discoverability. Both have their place in a content marketing strategy. But it’s nuanced. There’s no one-size-fits-all, there’s some middle ground, and you might need a bit of everything.

Note that Google doesn’t care if content is written by humans or with an AI. Their policy is: “Using AI doesn’t give content any special gains. It’s just content. If it’s useful, helpful, original and satisfies aspects of E-E-A-T (experience, expertise, authoritativeness and trustworthiness), it might do well in Search. If it doesn’t, it might not.”

Central to our work as an agency is to smartly craft content to meet budgets and deliver results. Here are three models to consider for content creation in an AI world.

Bingeable Content: The Quantity over Quality Model

This content is about surfing and snacking. Scannable reading that keeps users coming back for more with a steady stream of AI, SEO-performing, easily-generated content that generates clicks. Once called clickbait, the headlines often look like this:

• Top 10 Reasons to Ditch the Car

• The Truth About Chocolate Will Shock You

The content is generic and informative, often with a snappy headline that lures in the reader. As a bonus, it’s also optimized for search engines and based on commonly searched topics.

Karl is incredibly successful at generating this kind of content, voluminously, inexhaustibly, and much more productively than any human. And because the bar is lower, the concern here isn’t about brand differentiation. The goal is attraction – eyeballs on links that lead to traffic or clicks.

But one of the dangers of AI is it hallucinates; Karl says untrue things in an authoritative way. The AI was trained on a massive amount of information on the Internet, including Wikipedia. (Say no more.) So humans still need to be involved in editing and fact checking before publishing. And a lot more involved than most people think.

Who’s it Right For?
Bingeable content is great if your goals are low cost, high-volume content with clicks and fast delivery.

Brand Equity Content: The Quality Over Quantity Model

The polar opposite is content written by humans, curated like a Michelin-star meal. AI-generated text can only understand what others say about your brand, not the way you want it to show up in the world.

Humans create better stories and define a voice that makes you stand out, as only you can. The AI also gets repetitive, and it’s not always structured in a way that makes it readable. Karl is also not funny, not intentionally anyway.

Human-generated content still generates the highest SEO traffic. And fresh content, which hasn’t been published before, wins biggest when it comes to SEO results.

Because this content is unique, audiences are often won over by their longstanding affection and attraction to the brand. There is higher brand engagement. Only Nike can be Nike. Only Dollar Shave Club is Dollar Shave Club. Dove is unequivocally Dove.

A little test. I asked the AI to write a paragraph about what makes Art & Science different.

Karl’s Version

Art & Science sets itself apart from other agencies with its one-of-a-kind blend of creativity and data-driven strategy. It’s not your typical agency! They really dive deep into the sweet spot where artistry and technology meet, making sure design and analytics work together like peanut butter and jelly to whip up powerful, results-driven solutions.

Human Version

In 2013, we set out to build a new kind of agency, one focused around the exciting technology that was opening up new ways to move brands forward. Fast forward (and it has moved quickly!) and we have now created result-generating and award-winning digital experiences and campaigns for some of the biggest brands out there. And we’ve also invented some pretty cool shit along the way — come visit and check out our Museum of Awesome Things.

The obvious takeaway is Karl doesn’t have our genuine, brainy, scrappy voice. What it produced is hollow and trite. It bogs down in cliches like “wear multiple hats.” But it did generate its paragraph faster and definitely cheaper. (Note: there are already AI solutions in the market trying to ape the human voice, so it’s a matter of time before this gets more authentic.)

Who’s it Right For?
Brand equity content is right for brands that rely on ownable content that taps into the human experience, loyalty and authenticity.

Branded Binge-Worthy Content: The Middle-Ground Model

Here’s the nuance. What if it was possible to have both: a prolific AI that can churn it out with the emotional intelligence of a human?

With this model, the human writer is leading with the AI as a robot expeditor.

Generative AI would create a list of topics that people most often search, not unlike how we currently use SEO research at Art & Science. Then the human writer picks a few topics and puts the company’s lens on them. Instead of something generic, it becomes uniquely special.

For example, AI-powered tools can generate content outlines, research information, and even suggest creative ideas, streamlining the writing process and enhancing productivity. AI can also provide writing recommendations based on your brand’s preferences, helping to produce higher-quality content with greater ease.

That paragraph above was written by Karl. Was it noticeable? I made three quick edits to bring it in line with the voice of the rest of this article. But it’s not bad. It’s also functional material that I could have written, but it was faster with him.

Note that prompting the AI is an art. Often the copy needs a few rounds of refinement. There are many blogs and tips on how to prompt well.

Who’s it Right For?
Brands can hit their targets of mass content creation at a lower cost, without sacrificing individuality. It also hits the mark for ranking highly by Google and driving traffic.

Fusion in Content Marketing

The future of content marketing promises to be a fusion of technology and human ingenuity – the bottomless bag of delectable chips and the curated meal. AI is a powerful tool in this process. It’s getting better all the time.

The three models are not meant to suggest that marketers should pick just one. Your goals will drive how AI-generated content can be served up to support your business needs, brand and budget. You will likely need a bit of everything. Let us know if we or Karl can help.

This brand-equity article was written by a human, with Karl assisting on research and simple paragraphs. The human time invested was 4 hours.

Alison Butlin

Director, Strategy and Content