AI for Content Marketers: Friend or Foe?
No technology has been more hyped in recent years than Artificial Intelligence (AI). While observers and news articles like to pontificate on general AI as the proverbial sentient artificial lifeform, the reality is AI works best when applied to narrow, albeit numerous, use cases.
AI has the potential to be a transformative technology in various fields, capable of doing tasks humans find tedious and others simply better.
So what does this mean for content marketers? Should they be scared for their livelihood or excited about the possibilities? We believe content marketers should view AI as an opportunity. Effectively, AI can make the content marketer more effective.
This doesn’t mean content creators won’t need to continue developing their skills in the future—they will. But AI can augment those skills, perhaps turning them into superpowers in understanding audiences better in the name of delivering content volume, velocity, and variation like never before.
At WordPress VIP, we’ve already started to incorporate AI into our efforts to help content marketers be more effective and efficient, and it’s just the beginning. In this article, we discuss five ways content marketers can use AI within the framework of WordPress VIP Content Management and Content Analytics to save time and improve their work.
AI use case 1: Automatic classification
To be effectively created, structured, distributed, and measured, content needs a taxonomy.
One key element of content taxonomy is subject-matter tagging. Essential as it is, it can be burdensome, particularly when you need to decide in advance all the topics to write about in the next month, quarter, or year. That’s a big ask. No one had a content tag ready for COVID-19 in 2018 or one for Silicon Valley Bank at the start of 2023. New topics germinate daily in every industry.
To address this, you can leverage AI to discover new topics and automatically tag content that addresses each. Parse.ly’s Smart Tags does just this—automatically detecting and tagging topics in your articles, which increases the scalability, accuracy, and depth of your taxonomy. The underlying technology behind this sub-discipline of AI is called Natural Language Processing (NLP). With NLP, content marketers can automate the process of tagging their content, saving time and increasing reporting accuracy.
AI use case 2: Content recommendation
Keeping readers engaged and on site is a never-ending challenge for content marketers.
Determining the right article to suggest next is key. But manually recommending related, engaging content for every reader based on what they’ve just read would be outrageously time-consuming. It requires looking at (and understanding) massive amounts of data and different journey paths that change constantly as topics trend.
It’s another great use case for AI—and where Parse.ly’s Content API comes in. The API recommends high-performing related content on your pages and articles, even if the topics are phrased differently. This difference is key.
For example, an effective content recommendation engine needs to understand that “King,” “Monarch,” “Charles,” and “Royal” may all mean the same person. Simple pattern matching isn’t up for the task. But AI technology called transformer models are. They work to understand not just the words but the context and meaning in content.
The newest version of our content API, which utilizes transformer models, is 25% more effective than previous iterations that didn’t. By leveraging transformer models, content marketers can improve the accuracy of their content recommendations, leading to higher engagement and better user experiences.
AI use case 3: Text generation
The need for volume, velocity and variation is endless.
In our Content Matters 2023 Report we found 61% of content marketers want to create more content. But 44% listed resources, and particularly lack of staff, as the biggest challenge to doing so. Given current resourcing constraints, it’s hard to generate all the content you need, especially on niche topics.
AI can help with this as well. With text generation, you can create a paragraph on-demand, based on a prompt. It may not always be 100% on point and will likely require editing. However, it can get you a long way toward the goal in a fraction of the time.
The underlying technology behind this feature is natural language generation (NLG), which uses deep learning algorithms to generate human-like text. ChatGPT is the most popular and buzzworthy NLG technology. We currently have a capability in beta that puts the power and convenience of ChatGPT to use within the block editor in WordPress VIP. It has the potential to be a real time saver for content creators. (As an aside, we used ChatGPT to generate the first draft of this article, a fine way to kickstart things. But then came the heavy editing, fine tuning, and wordsmithing.)
AI use case 4: Image generation
Images are essential components of articles, but always pulling imaging from the same, limited library can result in bland, repetitive results.
Creating bespoke imaging manually can also cause bottlenecks, as graphic design talent and time come at a premium.
AI-driven Image generation can help close this gap and focus your designers on unique brand elements rather than simply generating accompanying images for articles. WordPress VIP has introduced an image generation feature (also in beta) that integrates the deep learning algorithms of DALL-E into our CMS to generate images on-the-fly, based on simple prompts.
AI use case 5: Getting help yourself
Up to this point, we’ve discussed technologies that can be integrated with a CMS to make content marketers more effective at their job. But can AI help content marketers be more effective in using the CMS in the first place? We believe so.
We know that reading documentation can be time-consuming, even frustrating. Though we pride ourselves in the ease of use of our tools and the quality of our documentation, we know we can always make both easier. Right now, with that in mind, we’re working on a way to ask a documentation chatbot questions and get instant responses. The underlying technology behind this feature is once again natural language generation and ChatGPT. Stay tuned.
Next in the AI realm for WordPress and content creators (and developers)
These are just five ways Artificial Intelligence can make content marketers more effective. We’re also exploring AI-powered blogging prompts, pattern assemblers for site content, and accessibility and usability improvement tools. Plus ways to help developers code better or automatically generate plugins, tests, and other tools. The potential is endless.
In summary, don’t look at AI as purely technology for technology’s sake, and certainly not as a threat. AI has the potential to solve challenges you face daily. Embrace it, partner with it, and use it across these use cases and others to do your job better.