You’re Not “Bad at AI”: You’re Just Missing a Few Essentials
I resisted AI for longer than I’d like to admit. When everyone at my networking group was buzzing about ChatGPT, I was the one quietly rolling my eyes into my coffee. I was a writer. A real one. The last thing I needed was a chatbot stepping on my craft.
Then I tried it because… well, curiosity got the better of me. And for a while, I thought my skepticism was completely justified, because what came out was exactly as soulless as I’d feared. It sounded generic and flat. It was missing even the faintest hint of personal style.
But here’s what I eventually figured out as I kept experimenting: the problem wasn’t AI. It was how I was using it.
I like to think of it like that time I got my first iPhone. I was so excited to use it, but it was so different from my flip phones that I spent days (literally) just figuring out how to send a text message or call someone. To be honest, I’m still learning how to fully use an iPhone 15 years later. But now that I understand what it can do and what I want from it, I know how to work backwards to get those results.
The same rules all apply to AI. The capabilities are there, but the user determines what level of quality comes out. And that’s not about being tech-savvy; it’s about setup and learning.
Now that I know what to do, I also know what not to do. After working with dozens of business owners on their AI-assisted content, I’ve seen the same five mistakes come up again and again. If your content sounds like a robot wrote it, you’re probably struggling with one of these.
Mistake #1: Prompting for Content Before Giving AI Any Context About Your Business
This is the big one. Most people open a chat window and type something like, “Write me a LinkedIn post about my coaching business.” Then they’re surprised when what comes back sounds like every other LinkedIn post they’ve ever scrolled past.
AI doesn’t know you. It doesn’t know your clients, your voice, your positioning, or what makes your approach different from the fourteen other coaches in your city. When you give it nothing, it fills in the gaps with the most average version of everything. It’s just like bringing in a fresh intern, sitting them at a desk, and asking for output without any real input.
The fix is foundational messaging. This is not a two-sentence description of your business. It’s a 35-page document that gives someone (or AI) the full picture of your business: who you serve, what they’re struggling with, what you believe, and how you talk about all of it. Once that exists, AI can work from your foundation instead of inventing one.
This is exactly why foundational messaging is the first step in Lean Marketing Lab — not because it’s a nice-to-have, but because without it, every prompt you write is starting from scratch.
And even better, in the program, we teach you how to make sure you won’t need to upload this doc (or several other essential foundational documents) every time you sit down to work with your AI.
Mistake #2: Asking AI to “Write a Blog Post” Without a Structure or Goal
“Write a blog post about the importance of consistent marketing” is not really a prompt; it’s more like a suggestion. An AI can take that in seventeen possible directions, most of them fine, but none of them exactly right.
Great prompts tell AI what the post needs to accomplish, who it’s for, what the reader should feel or do at the end, and how it should be structured. The more specific the input, the more specific the output. If you want an introduction that opens with a relatable scenario, a middle section that walks through three practical steps, and a closing CTA that points to a specific offer, say all of that. Don’t make AI guess.
Think of it like briefing a new writer on your team. If you hand them a topic and walk away, you’ll get something that technically covers the subject but misses the entire point. When you sit down with them and explain the angle, the audience, the goal, and the tone, you’ll get something you can actually use.
As an added benefit, having these prompts ready to go in a prompt library and using them over and over will ensure you have consistency in everything you create (social posts, blogs, newsletters, etc.), and consistency builds trust.
In Lean Marketing Lab, you get access to our full prompt library with all the instructions for writing everything from website pages to email sequences. These are fill-in-the-blank prompts, so you always get customization, but without rethinking or rewriting the whole prompt every time.
Mistake #3: Accepting the First Draft
On our team, we know the value of having multiple sets of eyes and multiple drafts for anything we write or create. The first draft is ALWAYS a starting point, not a finished product. This is true of human writing, and it’s especially true of AI writing.
If you see AI as a full replacement for your own work, you’ll get AI-quality work. This usually leads to content that’s technically fine but doesn’t quite sound like them, doesn’t quite land, and over time, creates a subtle inconsistency that erodes trust with their audience.
If you see AI (and use it) as a resource to get you 80% of the way there, you’ll get personalized, high-quality content, but without all the energy and effort it takes to get it to that first 80%.
Here’s what you should always do: treat the first draft as raw material. Read it critically and ask AI to revise specific things, or make the revisions yourself.
Here are some critical revision prompts you can use in an AI conversation:
“The opening feels too formal. Rewrite it to sound more conversational.”
“The third paragraph is vague. Can you make the example more specific?”
“The call to action doesn’t match the tone of the rest of the post. Try again.”
You can even ask what the AI would improve.
Each revision prompt is a chance to pull the content closer to your voice and your standards. Remember, the goal isn’t to get a perfect first draft. The goal is to know what to ask for next.
In our Lean Marketing Lab prompt library, we actually have a list of “revision” prompts to help the tool analyze the work and determine what needs to change. This gets you even closer to a finished product, so you can fine-tune and get it out into the world.
Mistake #4: Letting AI Choose the Tone
If you don’t specify tone, AI defaults to a kind of pleasant, professional neutrality that belongs to no one in particular. It’s not offensive or bad, but it’s not you, and your audience will feel that, even if they can’t name it.
Tone is one of the most important things to define in a prompt, and it’s also one of the most commonly skipped. “Professional but warm” means something different to AI than it means to you. The more specific you can be, the better: “direct, a little dry, no cheerleading, gets to the point quickly” is a much more useful instruction than “professional.”
This is another reason foundational messaging matters. When you’ve documented your actual tone, the words you use, the phrases you avoid, and the way you open a sentence, you can give AI a real reference point instead of a vague direction.
Mistake #5: Using AI for Strategy Instead of Execution
I like to think of AI as the “middle ground” for writing. It can help with ideation, execution, and revisions, but if it’s not working from a strategy, it’s probably not going to work very well.
If you don’t have a strategy, the content you get from AI won’t be grounded in your business goals, your audience’s actual questions, or the specific gaps in your content ecosystem (meaning the interconnected network of content that’s supposed to be moving people toward working with you).
Start with a strategy. You can ask AI to help you build it, but this needs to come from your knowledge of your business. Who are we talking to? What are we trying to accomplish here? How can the content you create work together to build a pathway for your audience from the visibility stage to the purchase stage?
Here’s where AI helps with the workflow after that:
- Ideation (i.e., “what are some topics or question people commonly ask about my work?”)
- Organization (i.e., “create a content calendar that includes…”)
- Briefs to work from (i.e., “outline this topic and ask me questions about it so I can provide you with specific examples and stories”)
- Revisions (i.e., “what could improve this post? How can we better connect it to our ideal audience?”)
- Repurposing (i.e., “take this blog post and create 5 unique social media posts from the content”)
Again, you’ll need to use your own judgment, and a real strategic framework, to decide what gets made and why. This is not a limitation of AI; it’s just an accurate understanding of what it’s for.
Your Prompts Are Only as Strong as Your Foundation
The thread running through every mistake on this list is the same: AI produces better content when you give it better input. Better context, better structure, better tone guidance, better goals—these are what will give you content you’re more than happy to post.
The business owners who get genuinely useful output from AI aren’t necessarily more tech-savvy than you. They just have clearer foundations to work from, and they’ve learned to treat AI like a skilled contractor: capable of excellent work, but only when your highly specific processes are working behind the scenes.
If you’re ready to stop guessing at prompts and build a content system that actually sounds like you, Lean Marketing Lab is where that starts! Reach out to us and let’s talk about what you can achieve in the Lab.
