I’m going to show you how to get better at writing prompts for ai.
We’ll use a simple prompt structure to help you skip the guesswork and get consistent, reliable results faster.
By improving ai outputs with this method, you no longer have to babysit the AI.
This is a big deal if you are an entrepreneur, marketer, or anyone who wants to use effective prompting for better content and strategies.
Many people throw half-formed requests at AI or give it almost no detail.
Then they expect perfect results.
They end up fixing errors for longer than it would take to do it themselves.
“If your prompt is unclear, your output will be too. Solid structure is the fastest fix.”
I’ll break this down into three simple lessons so you can fix your prompts and get better results right away:
Why Most AI Outputs Fail—and How to Fix It
Most people fail at improving ai outputs because they give vague prompts or miss key details, leaving AI to guess[1].
The 4-Step Formula for Prompt Structure
A clear prompt structure—defining a role, adding context, giving a direct command, and allowing clarifications—makes it easier to learn how to get better at writing prompts for AI.
Choosing the Right AI Model for the Job
You’ll get better results when you match your task to the right AI model because some tools are better at data, others at creative tasks, making writing ai prompts far more effective.
Instead of reading, would you rather listen to this as a podcast or watch it on YouTube?
👉 Listen Now | Watch Here
Why Most AI Outputs Fail—and How to Fix It
This is where most users go wrong.
They either give AI too little direction or overload it with too much information.
They treat AI like a mind reader, expecting it to fill in the blanks without clear instructions.
The result?
Confusing, generic, or useless outputs.
It’s not the AI’s fault.
The problem starts with you, the person writing ai prompts without structure.
When you feed AI a vague or cluttered prompt, you force it to guess what you really want.
That’s why so many people feel frustrated when AI “doesn’t work.”
“Most people don’t know how to write a good prompt. They’re either too lazy to provide clear instructions, or they expect the AI to figure it out.”
The fix is simple.
If you want better results, you need to stop treating AI like magic.
Instead, treat it like an assistant who needs clear, step-by-step guidance.
This is where the right prompt structure comes in.
Without structure, the AI will continue to fail you.
✅ What You Can Do Today
If your AI outputs feel random, start by looking at your prompts.
Ask yourself:
- Am I giving too many tasks in one prompt?
- Am I leaving out important information or context?
- Am I expecting the AI to guess what I want?
To fix this, write one focused prompt at a time. Remove extra requests. Be clear about your goal.
This simple shift is the first step toward effective prompting—and the fastest way to start improving ai outputs immediately.
The 4-Step Formula for Prompt Structure
There’s a simple formula that fixes this problem.
It’s the best prompt structure for anyone who wants stronger, faster answers[2].
You’ll hear many prompt writing tips online, but this formula is the one that actually works.
Step 1 – Define a Role
“You’ve got to tell the AI what role you want it to play.”
The role is like a job title.
“The role is what they are, what their job title is, what their experience is.”
For example, you could say, “You are a senior marketing executive with ten years of experience in content creation.”
That role helps the AI know how to answer.
“It creates a lens, a viewpoint for the AI to look at the information through.”
This is one of the most overlooked parts of prompt crafting.
Without a role, the AI doesn’t know how to frame its reply.
Step 2 – Provide Context
“You need to tell it the background, the specifics, the situation.”
Context is essential in good prompt creation.
It tells the AI what’s happening, who it’s for, and why it matters.
“It’s like briefing a real person. You wouldn’t give someone a task without telling them why or who it’s for.”
Without this, AI fills in the blanks on its own—and that rarely ends well.
Step 3 – Set a Direct Command
“You need to give it a clear, specific task. Don’t ask it to do five things at once.”
This is one of the biggest mistakes people make when learning how to write the prompt for AI
They cram too many requests into one sentence.
“When people give a prompt, they often give it five different tasks in one sentence. And it’s no surprise that the AI gives them a messy answer.”
Direct, focused commands lead to clear, useful outputs.
That’s the core of prompt optimisation.
Step 4 – Invite Clarifying Questions
“You want to give the AI permission to ask you questions before it answers.”
This is one of the most powerful steps in good prompt design, but most people skip it.
By saying, “What else do you need to know before you answer?” you let the AI fill in the blanks without guessing.
“It’s like a conversation. You wouldn’t expect someone to do a job without being able to ask questions.”
This small step improves effective prompting dramatically.
This simple, clear structure is the answer to the question of how to get better at writing prompts for AI.
It’s not about using fancy language—it’s about clarity, role, context, and one clear task at a time.
That’s the real secret to prompt crafting and high-quality outputs.
✅ What You Can Do Today
Use this formula every time you write a prompt:
- Define a Role — Tell the AI who it should be.
- Provide Context — Give it background and details.
- Set a Direct Command — Make one clear request.
- Invite Clarifying Questions — Allow the AI to ask before answering.
These are the most important prompt guidelines you’ll ever need[3].
Following them will improve every AI task you do—from writing to research to strategy.
If you want to master how to write the prompt for AI, this is where you start.
Choosing the Right AI Model for the Job
Another reason people struggle with improving ai outputs is because they use the wrong tool for the task.
It’s like trying to hammer a nail with a screwdriver.
No matter how well you follow the best prompt structure, if the AI model isn’t suited for the job, you’ll get weak results.
“If you’re not using the right AI for the job, you’re setting yourself up to fail.”
Not all AI tools are the same.
Some are built for writing, others for design, data, or video.
Choosing the right model is part of learning how to get better at writing prompts for AI.
“Different models have different strengths. Some are good at content creation. Some are good at data analysis. Some are better for coding tasks.”
For example:
- If you need help with writing content, tools like ChatGPT and Claude are built for that.
- For visual content, MidJourney and DALL·E are great at generating images.
- If you want to create videos quickly, Synthesia and HeyGen can help.
- Tools like Copy.ai and Jasper focus on marketing copy.
- Descript is perfect for editing audio and video content.
“It’s not just about the prompt. It’s about matching the prompt to the right tool.”
However, there’s also a clear and simple rule to start with.
“90% of the time, ChatGPT is a great start. It may not always be the best finisher, but it is the best place to begin.”
Choosing the right model also affects how much information you can include, how creative or technical the output will be, and how well it follows your instructions.
Every tip here is inspired by conversation with Jonathan Mast (AI Consultant and Coach) on the Mastering Tech Growth podcast.
If you enjoy listening or watching, tune in to the Mastering Tech Growth podcast on your favourite pod platform or catch it on YouTube for even more in-depth insights!
- [1] 5 Common Generative AI Prompt Writing Mistakes (And How To Fix Them) https://www.forbes.com/sites/bernardmarr/2024/10/01/5-common-generative-ai-prompt-writing-mistakes-and-how-to-fix-them/
- [2] How To Write Amazing Generative AI Prompts https://www.forbes.com/sites/bernardmarr/2024/09/25/how-to-write-amazing-generative-ai-prompts/
