AI and Headless CMS

While machine learning has been around for many years, the emergence of ChatGPT ( a chatbot developed by OpenAI) at the end of 2022 has led to AI becoming mainstream.

Ever since, many SaaS applications have introduced AI capabilities into their offerings. In part, it is to take advantage of the capabilities, but also to avail of AI ‘multiple valuation dividends’.

In short, the investment community has doubled down on AI and any reference to it has attracted a premium, acting as another powerful incentive to expedite its adoption.

Regardless, AI is here to stay, and it will increasingly weave its way into our workflows.

Content Management Systems (CMS) are no different.

So what are the areas that CMS have focused on when it comes to adding AI capability?

This short article explores some of the broad areas where AI and CMSs converge and seeks to assess the usefulness of the applications.

Content Creation 

Content creation (or generation) has emerged as a key area for AI applications. 

However, for many, this is a use case that can be problematic, to say the least. Students submitting AI-generated essays, and job applicants submitting AI-generated cover letters represent two examples where the use case is considered inappropriate. 

The same applies to web-generated content, be it blogs or otherwise. 

After all, the ease (and speed) with which AI can create content offers serious efficiency gains. However, the accuracy of the content is suspect, to say the least, given that the very essence of AI-infused content is regurgitating content pulled from elsewhere with no checks and balances as to its veracity. 

I for one hit the ‘back button’ whenever I encounter a piece that is written by AI (the signs are easy to spot). 

It also represents a risky strategy, as many believe that AI-generated content will eventually fall foul of Google’s algorithm for SEO - a risk that many want to avoid.

A broader issue specific to all CMS, is that much of the content creation process currently takes place outside of the CMS, be that in Google Docs, Notion or MS Word. Hence content gets into the CMS via ‘cut and paste’ rendering the feature within the CMS almost redundant. 

In short, the jury is still out on whether or not a ‘content creation’ AI capability is a valuable addition to a CMS or not. For some, it is an invaluable time-saving enhancement which rewards quantity over quality. For others, it serves to dial up the noise, diluting the signal-to-noise ratio and negatively impacting all content creators as the volume of content created grows exponentially.

Of course, content creation comes in many guises and also includes multimedia i.e. images and video (with editing capabilities sitting alongside origination). These lag behind text but are also gaining in popularity. Again the quality of these has improved in recent months. 

Perhaps I am old school, but for now, I continue to see value in original imagery compared to AI-generated content. But as with text, there is undoubted value in some of the other AI capabilities I delve into further below e.g. post-production enhancements.

Content Enhancements

Augmenting, enhancing or editing content sits in a slightly different camp to the creation process. This could range from spelling suggestions, and grammar improvements, to analysis of tone, through to summarizing larger pieces for subsequent use in socials. Sentiment analysis, labeling and categorization also represent augmentations you’ll encounter.

These enhancements make sense. They are suggestions that are designed to improve an existing piece of content, complementing the work of the content creator rather than seeking to act as a substitute. Again as before, content includes multimedia and we are all familiar with the powerful editing capability of our own smartphones.

Like all AI elements the key is to ensure it operates as a co-pilot rather than as a direct substitute. Hence, suggestions and recommendations can be considered and accepted/disregarded as the author sees fit.

SEO Enhancements

SEO enhancements sit outside of the content production/enhancement piece as a distinct AI feature, as they aim to ensure that the produced content has the constituting elements required to ‘rank well’ in search engines (primarily Google). 

In some ways, it is a natural evolution from WordPress plugins like Yoast that incentivise you to follow a recipe, which delivers green ticks as you hit certain pre-configured criteria. 

SEO AI support can range from flagging word count limits to the use of designated keywords in URL slugs, meta titles, meta descriptions and H1s. A more advanced AI lead approach tries to support the process, and also includes image SEO, an often neglected part of SEO efforts.

Again, I’d argue that these enhancements are worthwhile and can be considered before accepting.

Localization/ Translations

Another area where AI can offer value is in text translation. With this, there are no issues around hallucinations or factual misrepresentation - after all it is inputting content from one language and translating it into another. 

However, any AI translations should be flagged as such so the reader knows they are AI-generated, and should allow manual human overrides. After all, language is nuanced and context is everything, and AI can get some translations badly wrong. 

Personalization

Headless CMS are well suited to AI in part due to the API lead architecture, but also due to the structured content approach in operation, where there is a clear separation between content and code. This separation makes it easier to feed different content depending on different criteria e.g. serving different content to someone ‘logged in’ compared to someone who has not ‘logged in’.

Headless CMS will benefit from AI to offer more personalized and dynamic content delivery across various platforms, making it smarter and more responsive to user behavior.

— Source: Headless CMS: The Epic Battle With Monolithic Platforms, Tim Marsh, CMS Wire

Compliance

Compliance represents another area where AI can play a meaningful role - again as a support actor. Compliance is a laborious and time consuming task, and of course the increasing volumes of content being produced exacerbates the challenge. AI can help review written content, looking for anomalies which are then flagged for human review.

As Content Strategist, Michael Andrews claims:

Generative AI doesn’t replace writers or compliance reviewers. But it can help make the process smoother and faster for all by spotting issues early in the process and accelerating the development of compliant copy. While GenAI won’t write compliant copy, it can be used to rewrite copy to make it more compliant.

— Source: Supporting content compliance using Generative AI

While there are undoubted efficiency gains from leveraging AI in the broader content management space, there are also undoubted pitfalls. As we’ve outlined above, some AI-supported tasks work really well with AI input whereas others stray into areas where the use can be problematic e.g. externally facing content masquerading as “authoritative thought-leadership” content. 

We are very much in the early days of mass AI deployment, and as I’ve argued above the danger is that features and capabilities get rushed out without due consideration being given to any potential downstream consequences.

A starting point has to be the clear distinction between AI as a complement and AI as a substitute. With the former, AI can support tasks but as input for consideration, rather than delivering end-to-end tasks as a like-for-like substitute e.g. here is a full article.

It is also worth noting that there is so much AI innovation, that it takes some time to filter through the use cases, and that in the months and years ahead we are likely to really enjoy the benefits of it, as the most valuable examples and use cases become more well known. 

It is also worth restating that Headless CMS are very well suited to AI, the API lead architecture is designed to integrate seamlessly with other applications, with negligible vendor lock-in thus future-proofing your tech stack and giving you the flexibility to adapt to future developments.

The world of Artificial Intelligence (AI) is rapidly evolving, reshaping industries and transforming how we interact with technology. One of the key advancements in this domain is generative AI, which relies on robust, structured, and labeled data.

— Source: The Unseen Backbone of AI: Structured Data and Headless CMS by Bradley Taylor

So what are some of the benefits of using AI with your Headless CMS? The general benefits from AI include:

• Increased efficiency (reduction in repetitive tasks)

• Time-saving

• Reduction in errors

Adding these together results in productivity gains which are particularly welcome in a time-pressed marketing (or content) function. 

As mentioned previously it is worth noting that many of the tools used within the content creation process for many writers sit outside of the CMS. Be that Otter.AI to transcribe conversations, or Grammarly to improve the quality of the copy. Dedicated SEO tools like SEMRush and aHrefs are also adding AI capabilities.

In summary, the use of AI has exploded in recent months. Many Headless CMS are rushing to bake in AI capability. For many, it is still a case of ‘wait and see’ though. After all, they are enjoying much of the functionality in the tools they use sitting outside of their Headless CMS, already. The real breakthrough will be when these very same users commence the writing process within their CMS, putting an end to ‘cut and paste’  once and for all, and viewing AI as an assistant rather than a substitute.

I’ll leave the last word to Scott Kubie, Director, Content Career Accelerator, who advises companies to thread more carefully when it comes to adopting AI (especially for content creation).

But many other companies, maybe yours, are trying to run with AI before they’ve learned to walk with content operations.  The risk of tripping and slamming face-first into concrete — or a Canadian small-claims court — is only going to grow for companies laying off content experts, UX researchers, technical writers, and others skilled in making sense of complex information spaces. I expect to see more stories like this in the coming months and years, not fewer. But smart companies can avoid the worst of it by giving a damn about digital governance, content operations, and content design. And by hiring, or not losing, the people who care about that stuff, too.