How AI is speeding up - and slowing down - content creation

How AI is speeding up - and slowing down - content creation

It's been a year since ChatGPT turned the world of content creation upside down. After 12 months of testing and trialling various artificial intelligence (AI) technology, marketing agencies have not only got to grips with the tech, but many have incorporated it into the work they do daily. One of the biggest wins hailed by the marketing world is generative AI's ability to significantly speed up digital content creation. But where there are time savings, there are time losses too. Here, we take a look at the ways in which AI both speeds up and slows down the creation of quality content.

Slowing down content creation

The never-ending prompt

From what we're seeing, AI isn't removing jobs in the creation of written and visual content - and a lot of this is down to prompts. At the Power of AI, held by Campaign and Performance Marketing World in November 2023, Matt Cosad, Head of Data and Analytics at Kraft Heinz International, explained that their in-house marketing agency had grown as a result of AI, and much of this is because of the skills in prompting these models. Research predicts that by 2025 as many as 97 million people will be working in the AI space.

Understanding how to effectively prompt will have a huge impact on time savings. Knowing what prompt will get a certain result takes time and experience, and even training. We've all had that experience of putting in so many prompts that you come to the conclusion it would be much faster to just write the thing yourself. So, when we talk about the efficiencies AI brings, it must come with a caveat: providing the creative team member using the tool is well versed in the creative art of prompting.

The legal ramifications

Showing clients visuals of first-stage ideas is one thing, but getting a piece of AI-created content in front of the public is a whole other matter. And to be honest, one that's very unlikely. There are still so many legal concerns around AI and copyright that in-house legal teams are, currently, letting very little content reach the public-facing stage. So, discussions with your legal team must take place at the very earliest stage of any considered use of generative AI, otherwise you could find yourself at a point towards the end of a campaign, where legal steps in and puts a stop to the whole thing. Now that would be a lot of time wasted.

The post-production piece

We've already established that AI can be hugely helpful in speeding up the content ideas stage. If your legal team allows it, or if what you're creating is for internal use only, you may be able to use something AI has created and take it further. The issue is, while these systems are still in their youth, anything that AI produces still needs a huge amount of refinement. In our experience, AI does not yet provide a fully functioning end-to-end content creation tool. As a result, agencies looking to use AI to help create truly great content are finding that the time saved in pre-production is simply shifting to post-production. In all our excitement around the speeding up of initial processes, we're somewhat ignoring the fact that we're simply creating more work on the other side of initial asset creation. So, if we're viewing AI as an efficiency in the start-to-finish creation process, does it actually save any time at all? When AI gets to a point where it produces 'right first time' content, whether visual or written, this is when we'll be able to hail AI for its efficiencies.

The written piece

Written content was arguably one of the first areas people really trialled and tested generative AI in their marketing outputs - at Wardour, a content marketing agency at heart, it's certainly where we started. We even held an event in our offices on AI, where Wardour Founder and CEO Martin MacConnol said: “Generative AI can create a mass-produced sliced white bread. But it will never be able to create an artisan loaf." It's the artisan loaf our clients want. We believed it then, and we believe it even more firmly today, especially when it comes to written content. One thing AI simply cannot do well enough yet, in our opinion, is create a brand's tone of voice - it will never become a content creation service. No matter how good your team is at prompting these systems (Jasper’s new Company Intelligence tool, for example, allows users to upload positioning and strategy documents to create more strategically aligned content) content creators are pretty much having to re-write anything these systems create because of its ‘sameness’. It's pretty easy to pick out a piece of AI-generated copy that's made it into the public eye (we've certainly managed to spot the odd blog post and piece of social content). And in the latest Google trends (read about the Google helpful search update here), content that's clearly used AI to create SEO content that supercharges rankings will now do less well than it used to. So, despite all the progress with AI, we believe we'll always need to put time into creating truly valuable content, created by humans, for humans.

Where will time be spent next?

AI produces a huge amount of data that needs to be analysed before we can really refine how we use the myriad systems in better content creation. Measuring this data manually is now no longer possible, due to sheer volume, so it could well be that we see systems that, ironically, use AI to do this for us. With this data we should be able to understand what works and what doesn't (which will be particularly helpful when it comes to using prompts in any form of content creation service) and in doing so significantly speed up the time spent in getting the most accurate outputs, and in turn reducing time spent in post-production. The barriers around legal use remains to be seen…

But it's important to remember the power of the human touch in any content strategy and content creation. The human touch leads to engaging content. As Cosad said in the Power of AI in Marketing summit: “AI is still very much an enabler of our marketing, rather than taking over and driving the whole shebang. AI is going to be just another paintbrush in the hands of our creative teams. It's still going to have to be used alongside creative people. Combining it with people is always going to be very important.”

To learn more about Wardour's AI capabilities and how we use it in our marketing campaigns, email us at hello@wardour.co.uk.