Many professionals and business owners treat AI assistants like traditional search engines. They type in a short, vague phrase like "write an email to a client" or "ideas for a new product" and expect a flawless, ready-to-use response.
When the output turns out to be generic, robotic, or completely off-target, they get frustrated and conclude that AI is overhyped.
The truth is: AI tools are only as good as the instructions you give them.
I always tell business owners that writing a prompt is exactly like delegating a task to a human employee. If you hand a new team member a blank sheet of paper and say "write a business proposal"—without giving them any background context, target audience details, or quality guidelines—they will struggle, and you will be unhappy with the result. The exact same rule applies to AI.
By using a simple, structured formula, you can transform your AI outputs from generic fluff into highly valuable, professional business assets.
In this article, I will share the four-step prompt formula I use every day to get excellent results from tools like Claude and ChatGPT, along with simple prompting best practices.
The Four-Step Prompt Formula
To get precise results, every important prompt you write should contain these four elements: Role + Context + Task + Format.
Let’s break down how this works:
1. Define the Role (Who is the AI?)
Start by telling the AI what perspective it should take. This helps the AI narrow down its training data to focus on the right terminology and expertise.
- Vague: "Write a sales email."
- With Role: "Act as a B2B sales development expert with 20 years of experience in industrial packaging."
2. Provide the Context (What is the background?)
Give the AI the background information. Who is the target audience? What are the pain points? What are you selling?
- Vague: "It is for a new eco-friendly box."
- With Context: "I run a manufacturing company in India. We are launching biodegradable shipping boxes aimed at D2C e-commerce brands who want to reduce plastic waste and showcase sustainability."
3. Specify the Task (What exactly do you want?)
State the exact task you want the AI to perform. Be action-oriented and clear.
- Vague: "Make a cold email."
- With Task: "Draft a highly personalized cold outreach email introducing our new eco-friendly shipping boxes to a procurement head."
4. Set the Output Format (How should it look?)
Tell the AI exactly how to structure the response, the length, the tone, and what to avoid.
- Vague: "Make it short."
- With Format: "Keep the email under 150 words. Write in a conversational, direct, and professional tone. Do not use generic corporate buzzwords like 'synergy,' 'streamline,' or 'revolutionary.' Divide it into a clear hook, value proposition, and a simple call to action."
Combining It All: The Difference in Results
Look at what happens when we combine these four elements into a single prompt:
"Act as a B2B sales development expert. I run an industrial packaging firm in India. We are launching biodegradable shipping boxes for e-commerce brands. Write a cold outreach email to a procurement manager introducing our boxes. Keep it under 150 words, conversational, and direct. Do not use buzzwords like 'synergy' or 'revolutionary'. Highlight a simple call to action at the end."
By writing this detailed prompt (which takes less than a minute to type), the AI will generate a highly professional, ready-to-send draft. You will save 10 to 15 minutes of rewriting and editing.
Best Practices for Better Prompts
Here are three simple tips to immediately improve your daily interactions with AI:
1. Use "Few-Shot" Prompting (Give Examples)
AI learns exceptionally well from examples. If you want the AI to write a newsletter in your personal voice, copy one of your previous successful newsletters, paste it into the chat, and say:
- “I want you to write a newsletter about topic X. Use the exact same writing style, tone, and sentence structure as this example: [Paste your example here]”
2. Treat It as a Conversation (Iterate)
Do not expect the AI to get it 100% correct on the first attempt. Treat it like a feedback loop. This iterative approach is even more critical when working with different AI models to execute complex tasks. If the response is too formal, tell it: "This is too formal. Re-write it as if you are explaining it to a friend over coffee." If it missed a point, say: "Add a paragraph explaining the cost benefits of our solution."
3. Avoid the "Do Everything" Trap
If you have a complex project (e.g., creating a marketing campaign), do not ask the AI to write the strategy, build the emails, and write the social posts all in one prompt. Split it into steps. First, ask for the strategy. Once you review and approve it, ask: "Now, based on this approved strategy, let's write the first email."
Final Thoughts
AI prompt engineering is not a technical programming skill; it is a clear communication and management skill. As documented in the official OpenAI Prompt Engineering Guide, using clear instructions, providing reference text, and splitting complex tasks into simpler subtasks are foundational strategies for getting the best results. By practicing the Role + Context + Task + Format formula, you will unlock the true potential of these tools, saving time and driving better outcomes for your business.