AI For Content Creation: 3 Specific Use-cases and Examples

For GTM teams, content is one of the most critical things you have to get right and let’s face it, you’re never done. It’s literally how you “converse” with the market and if you break it down further, you need to converse at different milestones, with different messages and to different teams, throughout the journey. 

Suffice to say that content creation and actions can get pretty complex and it definitely needs proactive management.

What do we expect from our content?

Let’s review all the demands we put on it. Content must:

• Be inspiring and interesting to consume.
• Differentiate and convey the “why”.
• Tell a story and keep audiences coming back.
• Be highly relevant and not interruptive.
• Pick up from where it left off.
• Be available in many forms, written, audio, visual and interactive.

As a content marketer, likely working with the product team, plus growth and demand gen teams, you need to build a plan with a calendar, owners, themes, key topics and a host of keywords to make sure you’re optimizing for the coveted SERP rankings.

Why creating compelling content is hard

Because content is central to every campaign, you need to incorporate cycle times into your calendar and be fully prepared before launch, especially when you have pre-set KPIs with performance expectations.

Content feeds your ads across many channels and while ad copy is short, it takes many iterations to get right, and hit those expected conversions as you pour more channel dollars into bidding or other paid media programs. Suffice to say there’s a lot of moving parts with lots of deadlines and dependency alignment.

Content consumes so much time

Teams burn weeks every quarter creating and fine-tuning content. Even with lots of tricks such as writing once and producing for many media types, it burns precious hours. Even when teams outsource, it takes cycles to manage and review. And, you don’t always get optimal results because you’re constantly enabling and guiding.

We all love ChatGPT but there’s got to be more!

ChatGPT has become every marketer’s best-friend. We love it. So much more than a quick answer. You likely use it for deep insights, advice on frameworks and certainly as thought starters. It’s great when you’re facing writer's block. Or checking out what your competition is up to.

It’s a LLM model, becoming more commodity-like. There’s other LLMs, of course but it’s the most prevalent for content marketers. However, as a user, you still need to take time to review, cut and paste the useful parts into another tool and then apply context and decision making before execution. Sure, efficiency gains are made, but it requires the human brain to decide where to “deploy” the output.

As a ChatGPT user, you realize that you have to juggle with many tabs open, pulling from different sources. You need to then manage different prompts in different sheets. When you’re collaborating with a team-mate, it gets even more complex. Now you need joint sheets, manually working with different prompts and on you go updating. At some point you’ll have an asset library to manage in another tool and that becomes a full time job.

Content as part of an End-to-end Workflow

The context comes when you consider your audience, why it’s relevant and how to differentiate. Also, from a timing perspective when should it be delivered and at what cadence, especially in a series of messages going out.

An AI workflow is essentially a series of steps, or tasks that are automated for faster cycle times. ChatGPT’s value is just one step in the series. A workflow starts with a dataset or a structured list and then content is automatically generated, based on specific fields (i.e. context) from the input data. Once the AI content is generated, it’s reviewed and then the output can be used in numerous forms. 

Let’s take 3 quick use-cases to “create targeted, personalized content”:

Perhaps your team just launched a super-cool feature and you need to start amplying messages and get it into the right readers and viewers. You want a fast, efficient way to:

• Generate a LinkedIn message for specific personas and accounts
• Generate a 3-touch email message for specific persona(s)
• Generate content for a social post to a specific group or account

This end-to-end flow is more intelligent and helpful than a simple Q&A ChatGPT type tool. It considers relevancy, and is doing the tailoring of content, based on the recipient. It’s essentially “spoon-feeding” you what you need, taking in important considerations to execute the steps.

Readiness for Your AI Content Workflow

In many cases, all you have is the name of an account and some email addresses for the best-fit users. A quick, easy AI app can take those emails, parse and validate the company, its domain and even better, enriched by adding fields of useful context such as industry sector, employee size, HQ location, plus validate those emails. 

With data hygiene and context setting done, your content becomes more tailored.

‍

‍

Screen-capture: this AI App is taking your data, filtering, extracting company details and enriching with more info and relevant fields about the accounts. Save and download in a few minutes.

AI is great for highly specific content-related tasks

Of course you don’t want to go crazy out the gate. Start with small, valuable use cases and then grow from there. The more specific the task, the faster you learn and the more proficient you become.

Here’s a perfect example of one specific task:

1. Translate content

You want to translate a piece of content for a customer in a different geography. Sure you could mess around with Google translate and then copy/paste but that’s really hard to do at scale. It’s manual and time-consuming. It’s so much easier to use a pre-built workflow and let AI do the lift, all in one central platform that does lots of other useful AI tasks.

In this case, you upload the content and use the translate operator to do the work, save to a file and then use it in your preferred communication channel.

MarkovML has a pre-built template for this very task. No coding required. Easy, simple and secure.

It’s an easy win and what’s more you share with the broader team and train everyone on that specific workflow.

So, how can teams drive AI content use-cases and learn?

AI comes to life when the team’s creative minds work collaboratively to define scope, outline the important steps and then decide what to do with the output. 

Truly the only difference between using AI and prior manual approaches is you’re enabling AI to do the lifting. Without the team’s direct involvement, you may not be aligned on strategy and goals and also you may not be actively learning, which is critical.

AI Experimentation is great. However…

Embrace as one team to see the full value 

In a recent market survey run by Tangence, the things that slow teams down with AI show surprisingly “technical limitations” followed by “skill gap”. 

I’m guessing the “tech limitations” is more to do with integrations and features for end-to-end workflows, which are maturing crazy fast, everyday. Also coupling that with “skill gap”, often the team doesn’t have a good understanding of what’s possible with AI.

For MarkovML’s customers who are fully embracing AI for content creation, these are the general approaches we see over and over. I’ve condensed it down into 3 key considerations, as it’s easier to remember:

1. What content outcome do you want?

Think of it in the context of the audience and apply a step-by-step workflow approach. The easiest way to think about this is:

• What’s your data input (i.e. context)
• What task or action do you want to take (e.g. translate)
• What output result are you aiming to achieve? 

2. Will this workflow help other team-members and bring efficient scale?‍

• Can this workflow spawn new ideas and be further modified?

3. How will you measure overall performance and conversions? And, is your tracking setup correct?

These phases of getting any campaign off the ground are pretty commonplace but you do need to apply a new way of thinking or “lens” when asking what AI can do for you. 

Of course experiment with the tools and workflows before you start to scale and what’s more, measure the time it takes to execute. You will absolutely need to know that before-after picture for the full team’s learning and ongoing adoption of AI for content creation.

I’m always open to chatting with anyone on the GTM side of business operations. Our mission is to bring AI capabilities to all and make it easy to use, delivering game-changing value. 

Reach me at: pankaj@markovml.com