Table of Contents
With recent advancements in generative AI, there’s been understandable excitement among marketers for the potential of AI tools to replace human processes in content production. Opinions are spanning the spectrum; from those pronouncing that AI can already replace human copywriters to those pumping the breaks and pointing out its risks.
As with most divisive issues, the truth lies somewhere in the middle. The question that content marketers should be asking is, “how can I use AI to maximize the ROI of my time spent in creating content?” That’s the question that this article seeks to answer. Like nearly every marketer who makes content, we’ve been rapidly testing and learning how AI can augment production and we’re excited to share what we’ve found.
How Do We Increase the ROI of our Time?
The answer is simple. Reduce time spent on low-value-added (LVA) activities with the idea that it creates more time for high-value-added (HVA) activities. LVA activities are those tasks that yield little value (profit, fulfillment, personal growth, etc.) for the investment of time. HVA activities are just the opposite.
Here are some examples of LVA activities that most content marketers can relate to:
- Manual data entry
- Idle time spent moving between apps, systems, or modes of thought
And, HVA activities where most of us would love to maximize our time in:
- Strategic planning
- Performance analysis
- Process improvement
- Building relationships
- Skill development
Instead of viewing AI as a threat or a complete solution, let’s evaluate how it can be used to transfer time away from LVA activities towards HVA activities. In doing so, we can be of greater value to our organizations or clients (which, in turn, brings personal gain) and derive greater fulfillment from our work.
First, Let’s Acknowledge the Realities of AI Content
While there’s plenty to be excited about with AI, you have to understand its limits to use it most effectively. Here are the issues we’ve bumped up against most in our testing.
The accuracy of AI content is reduced by its knowledge cutoff
AI is trained on existing data and that data has a cutoff. For ChatGPT, it’s 2021. Meaning that the more fast-evolving a topic is, the higher the risk is for AI to give you outdated information.
AI can be wrong
When you give, say, ChatGPT a prompt, it doesn’t go looking for the answer to retrieve. Instead, it makes “probabilistic guesses about which bits of text belong together in a sequence” based on the data that it’s been trained on. Its guesses can be wrong, which can produce copy riddled with errors.
For example, here I asked ChatGPT to tell me the best places to stay in Nederland, CO and the response was way off:
AI content can be, and often is, repetitive in content & structure
From our testing, we’ve seen just how often generative AI tools can spit out copy that’s too formulaic. Granted, ChatGPT is not advertised as a copywriting tool, but we have seen similar results with ones that are, like Jasper and Copy.ai.
AI can’t speak from personal experience
When you read AI-generated content do you get the sense that it’s kind of… generic? That feeling likely comes from the fact that the “writer” has never experienced what it’s writing about. That missing human factor can make the copy rather bland. While this is perfectly fine for purely informational content, this issue becomes much more apparent when asking AI to write copy that’s meant to influence a reader.
AI might plagiarize
AI is trained on real-world data and may, as a result, produce copy that closely resembles content that already exists. In some cases, this could rise to the level of plagiarism. Another concern is that if multiple people enter the same prompt, they might receive nearly duplicate responses. So, content marketers who use AI content verbatim may unknowingly publish content that is also being published elsewhere.
8 Content Production Tasks You Can Speed Up with AI
Based on the limitations we covered above, it’s clear that generative AI is far from ready to be a complete solution for content production. But, it’s plenty powerful to alleviate time spent on low-value-added (LVA) content production tasks. Let’s dive into a few use cases:
1. Come up with content ideas
If you have a topic, ask AI to help you come up with content ideas. It’s no substitute for stakeholder interviews and competitive analysis, but it may help you surface obvious gaps in your content plan.
2. Develop first drafts of content outlines
The structure of your content is likely going to follow some logical ordering that AI can easily anticipate. Offload that first pass of your content outline to AI, then adjust as needed.
3. Discover FAQs to include in your content
“What am I missing here?” Use AI to help you quickly identify questions or subtopics that you may have forgotten to include in your content brief.
Bonus: Ask ChatGPT to write FAQ schema markup that you can include with your content when you publish!
4. Break through your writer’s block
For the reasons we covered earlier in this article, AI-generated copy isn’t perfect as-is and probably shouldn’t be blindly copy-pasted into your content. But, it can certainly help you work past stubborn writer’s block. Ask AI to write a few samples for your to get your creative juices flowing again.
5. Rewrite copy to fit your criteria
If you’ve got a sentence or section of copy you don’t like, have AI take a spin at rewriting based on your needs. It may save you time from having to scour thesaurus.com.
Here are some examples of when you can ask AI to perform rewrites:
- Shorten a lengthy explanation into something more concise
- Rewrite academic copy to have a more casual voice
- Convert British English spelling into American English spelling
- Substitute for a word or phrase you don’t like
6. Proofread and check for plagiarism
This is where a tool like Grammarly shines. Its AI capabilities can help you improve clarity, strike the right tone, rewrite subpar sentences, and even detect plagiarism.
7. Generate copy for certain use cases
When you’re using AI to help you save time writing copy, you’ll want to be sure not to negate those savings by spending more time editing and fact-checking. AI content requires fewer adjustments when it meets these criteria:
- Evergreen. If it’s a topic that changes rapidly then AI’s training data is likely outdated.
- Informational. The more straightforward the topic, the better AI will perform.
- Well-documented. The more niche the topic, the more likely AI is to be wrong.
In addition to body content, AI can be a big time-saver for writing metadata, coming up with headline ideas, or writing summaries of already-written copy.
8. Use the GPT for Sheets and Docs extension to reduce context switching
If you’re using ChatGPT to assist in your content production process, then I highly recommend you try out this extension. It saves you time from having to jump back and forth between the OpenAI website and your keyword research, content calendar, or content brief document.
I’ve found this extension to be a big time-saver for populating content planning spreadsheets with things like article headline ideas, sample meta descriptions, and topic-related FAQs.
Using this extension, I had ChatGPT write this next paragraph in the Google Doc I used to draft this article without having to click away:
Context switching involves the act of alternating between two or more tasks. At its most basic level, this means shifting from one task to another in order to complete a job. The act of context switching puts a strain on the brain, essentially overloading it with too much information at once. Our brains are built to focus on one task at a time, and when we attempt to multitask, we struggle to switch between different tasks quickly and effectively. As a result, context switching can lead to feeling overwhelmed and mentally exhausted.
You’ve probably noticed that most of the screenshots from this article are from ChatGPT, but that’s far from the only AI tool that can assist in content production.
Here are a few AI content tools we’ve found to be helpful besides ChatGPT:
Use AI to Augment Content Production, Not Replace It
Generative AI gives content marketers another tool with which to create content better and faster. However, in all the excitement of exploring AI, it’s important not to lose sight of the fact that your content has to work. It has to resonate with the humans reading it or else it won’t drive any value.
In our testing, we see a tipping point where overreliance on AI starts to degrade the quality of the content. It’s up to content marketers to balance these positive and negative effects in order to maximize the ROI of their content and their time.
If you’re looking for an outcome-driven content team to help you propel your search marketing performance forward with the right balance of AI and human input, drop us a line!