Maximize your content with ChatGPT, the 5 things you need to know

ChatGPT has become a staple of everyone's marcom content tools. Within just a few months, it wiggled itself in and became as common as Google for search. 

How can you maximize your content output while staying on-brand and authentic and getting your message across to your audience? 

How to elevate your content from mediocre to good to great while using AI and ChatGPT?

The 5 things to know to maximize your AI output

Before you dive into the prompt copy-pasting to get some ChatGPT output, it’s you who has to innovate and dare to challenge the status quo. If you want to make an impact, you still have to do the work. AI can help with insights, inspiration, ideation, research, and crafting the copy, including proofreading.  But it's still you to give your content a human touch and show that you care. 

On LinkedIn and every AI course, you stumble upon the templates, how-to’s, and advice, whether or not provided with copy-paste-ready prompts. It is all very convenient and easy – every monkey can do the job. Yet, if you know how to zoom out and use ChatGPT and AI as intended, you can give yourself an advantage and an edge in your content creation skills.

Here are a few things to know to get the most out of your AI collab.

1. Understanding ChatGPT's role:  is not an intern to boss around
2. Human and AI: a powerful duo, yet it’s your job to provide insights
3. Deep diving with ChatGPT; down the rabbit hole
4. The professional edge: ChatGPT-4
5. People recognize ChatGPT-created content

Let's dive into the details. 

1. Understanding ChatGPT's role

ChatGPT is not an intern you can boss around to do your dirty work. It’s a co-pilot; it's your wingman to help you do better.  

It’s a great sparring partner; it’s still you who has the brains and has to guide ChatGPT to get the desired output. 

Good to be aware of… 

• No new insights or strategies. By now, we all know that ChatGPT is a generative AI. In short, it’s a language model with a chat-like interface that works based on probabilities. This means ChatGPT doesn’t give you genuinely new insights, strategies, or ideas. How generative AI works 
• No bias or judgment. The best part of generative AI is that it has no bias, judgments, or view of its own. ChatGPT can help you by giving rational arguments. Together, you can sharpen your thinking, broaden your mind, and come up with a different approach.
• Conversational design. Bossing around ChatGPT doesn’t get you anywhere. As said, it’s you who’s in the driver's seat. Chatting with ChatGPT is where the magic is created. Notice that it’s already in its name.

2. Human and AI: a powerful duo 

It's you who writes the prompts based on insights and experience. It's AI that replies based on probability and predictability; AI only echoes what's already there. 

You and your mindset create the magic. A quick tip to get the most out of your collaboration is to use AI as your assistant and equal and not boss it around. You do that by being friendly and starting a conversation with AI, not giving an order. This does something neurologically in your brain. 

That’s why you should say ‘hi’, ‘thank you’, and ‘please’ to ChatGPT. It’s not that AI cares; it triggers your mindset on how your brain works. Credits to Conor Grennan. 

• ‘You are treating ChatGPT like a tool; that’s how we interacted with the internet for a long time. However, ChatGPT doesn’t behave like a tool; it behaves like a human. Therefore, to get the most out of the tool, we have to behave like we are talking to a human and not a tool. And start your prompt with ‘Hi’.’
• ‘Starting with ‘Hi’ is not just common courtesy; it does something neurologically in your brain. It establishes respect and rapport. This is social identity theory; this is the cooperative principle. If you treat the other person (or tool) as a partner, you are going to initiate a more productive dialogue, which is exactly where the power of ChatGPT lies.’

Example: my prompts

I love equality and collaboration over commanding. Next to ‘Hi’ and ‘thank you’, I also say ‘please’  and ‘can you…’ to ChatGPT. 

I like a more human conversational tone over the directive bitchy requests. This just does something to my mind, and I notice it improves not only my prompts but the output as well. 

3. Deep diving with ChatGPT 

If you stick to the surface, you'll get what everyone gets. You write a prompt for a 1000-word content piece on the ‘Reasons why not to use ChatGPT’, and you’ll get the same mediocre content as everyone else. 

Deep diving is where you can stand out and differentiate from others that use the same prompts. You can prompt your way into the rabbit hole and surface those details that others don’t.

• Understanding and context of the topic. You know your craft and have an in-depth understanding of the topic. You do the strategic thinking and provide the context. Gen AI uses existing content for its output. It’s an echo of what is already there. Echoing and repetitive content can be boring and not bring new insights or points of view. 
• AI has nothing new: echoing. Be aware that ChatGPT and AI have no new insights or unconventional thinking to broaden minds and horizons. As Gen AI works with probability and predictability, it brings nothing new. Using ChatGPT for content creation is echoing what's already there. You specify and drill down into the prompt to surface the magic and show that you care (and know).
• Inaccuracies and outdated. Also, be aware of inaccuracies, trustworthiness, sources, and up-to-date knowledge. Even though the recent update makes ChatGPT tap into the www for information, it can be trusted (yet). Not only in the sources, also in outdatedness and (cultural) context.

Example: my deep dives

I always start with a broad yet very detailed prompt to kick off the conversation. A prompt of around 50 to 100 words is my default. 

Once I get the first reply, I prompt-on-prompt and chain my requests together to dive into the details and get more out of it. During this conversation between me and AI, I continuously check on validity and whether it makes sense. I do the thinking; AI is my wingman. 

→ Tips to improve your prompts

If you really want to get the most out of ChatGPT, I strongly recommend upgrading to the paid ChatGPT-4. The difference is significant and worth the investment.  

4. The professional edge: ChatGPT-4

Since October 2023, I have been playing with ChatGPT-4, the paid subscription. I have to say, there are differences with the freebee version; I’m not sure if that’s due to the new releases or the paid version.

The most important is that it feels more accurate and to the point. There’s a little less fluff. And the output is in a different tone of voice—a little less bland and vanilla. Yet I can’t prove this; it’s my gut feeling.

It’s also faster. With ChatGPT3-5, you sometimes had to wait a few seconds; now, the output is almost instant.

Custom Instructions and build your own GPT

The best part is that you can use Custom Instructions to personalize your output. By using the Custom Instructions, my results went from good to great. 

Lastly, you can build your own GPTs to automate specific requests for proofreading your content and creating captions for social media and other prompts you use frequently.  Although it sounds and looks complicated, once you learn the trick, it only takes about 5 minutes!

Is it worth the $24.20 monthly fee? Absolutely. 

5. Recognizing ChatGPT content

We have been working with AI-generated content for over a year,  we know people can recognize it. It's something in the details; the words used, the length of the sentences, and a gut feeling that reveals it. 

Here are a few signs how to recognize ChatGPT content.

• Listicles. There is no real intro; there is an immediate dive into the listicle. The structure of AI written copy is almost always the same. You do the prompt, and ChatGPT provides the answer without context sketching or personal notes on why it matters. And the structure is always listicles or similar. Even if you remove the bullets, you recognize this structure in the text. ChatGPT has a catalog-like way of giving information. 
• Originality. Lack of originality and creativity. Gen AI relies on data patterns, probability, and predictability — nothing creative to spark joy and engagement.
• Not human, no personality. No human touch or cultural sensitivity. Emotions, nuances, irony, humor, and surprises are what make content stand out. That’s something ChatGPT can’t do. The output is based on data and lacks personality and connection; it’s too clean and has no bias.
• Not consistent. Lack of consistency. As there’s no personality, there’s no consistency between the content items. The only consistent thing is that the content produced by Gen AI lacks identity and is bland and vanilla.

You can fix this with a 3-step guide to combine AI and authenticity for amazing content. 

Wrapping it up: AI is amazing if done correctly

Don't get me wrong — I love AI. And it's amazing to create and improve content. The secret is that I give some human love and care to my audience and content, and I use AI along the way, not vice versa.

At the end of the day, it's your job to elevate your content and create something awesome. Show genuine effort and care in every piece you produce.

These are the five things to know to get the most out of your collaboration with ChatGPT: wingman, you have the knowledge and context, go into the rabbit hole, experiment with paid ChatGPT-4, and know that people recognize ChatGPT content. 

1. Understanding ChatGPT's role:  is not an intern to boss around
2. Human and AI: a powerful duo, yet it’s your job to provide insights
3. Deep diving with ChatGPT; down the rabbit hole
4. The professional edge: ChatGPT-4
5. People recognize ChatGPT-created content

Now it's your turn to apply all this and create some awesome content to make an impact. 

Stay ahead: subscribe to my newsletter

If you want to stay up to date on AI, content, and marketing based on insight and how it matters to you, subscribe to my LinkedIn newsletter. And you'll automatically receive my monthly update on the LinkedIn algorithm in your inbox. 

Two-click subscribe, and one-click to unsubscribe if you don’t like it.

Easy does it.

→ Subscribe