How to Write Better Prompts for AI Tools

AI tools are powerful, but their effectiveness depends largely on how you prompt them. Whether you're creating content, writing emails, designing a brochure, or generating social media posts — your results start with a well-constructed prompt.

1. Be Clear and Specific
   • State exactly what you want the AI to do. Avoid vague instructions, as they often lead to generic or off-target results.
   • Focus on a single goal or task per prompt. If you have multiple requests, break them into separate prompts.
2. Provide Context and Details
   • Give the AI enough background to understand your needs. This can include the intended audience, desired tone, or any relevant information about the scenario.
   • If you want the output in a specific format (e.g., list, table, email), say so explicitly.
3. Use Structured Frameworks
   • For complex tasks, outline the structure you want. For example, provide headings, bullet points, or even use JSON for character creation or data tasks.
   • Example: Instead of “Write an essay about automation,” use:
     • Title: [Your Title]
     • Thesis: [Main Argument]
     • Key Points: [Point 1], [Point 2], [Point 3]
     • Conclusion: [Summary]
   • This helps the AI organize its response logically.
4. Mimic Natural, Conversational Language
   • Write prompts as if you’re speaking to a person. Natural language is easier for AI to interpret and respond to accurately.
   • If the initial output isn’t right, converse with the AI: ask for clarification, request changes, or specify what to improve.
5. Use “Act As If” or Persona Instructions
   • Direct the AI to take on a specific role, such as “Act as if you are a marketing expert” or “You are a personal trainer.” This guides the AI to tailor its response to your context.
6. Give “Do” and “Don’t” Instructions
   • Specify what you want included or excluded in the response. For example, “Do include tomatoes, don’t include chili peppers.”
7. Provide Examples
   • If you have a preferred style or format, include a sample or describe it. This gives the AI a clear reference point.
8. Iterate and Refine
   • Don’t expect perfection on the first try. Adjust your prompt based on the AI’s response—add details, clarify instructions, or change the structure as needed.
9. Keep It Concise
   • While details are important, avoid overloading the prompt with unnecessary information. Concise, well-structured prompts are easier for AI to process.
10. Evaluate and Give Feedback
    • After receiving a response, tell the AI what worked and what didn’t. Continuous feedback helps refine future outputs.

Why Prompts Matter 🔝

A well-structured prompt helps AI understand your goals, context, tone, and output format. It’s like briefing a smart assistant — the clearer the brief, the better the output.

Start small, grow with interaction. If you’re unsure how to write a complete AI prompt, begin by stating a broad area. Your AI tool can guide the process — ask it to question you for clarity. Once the tool collects enough context, it can generate an expanded, well-structured prompt covering all key sub-topics.

For example, I was asked to introduce the concept of AI to college students. I started with a basic prompt:

 "Help me create an introduction to Artificial Intelligence for a college audience." 

To build further, I asked the AI to prompt me with follow-up questions. It asked:

• What is the expected duration of the session?
• Should it include examples or a demo?
• Is the audience from a technical or non-technical background?
• Do you want to cover applications like ChatGPT or just core concepts?

Once I answered these, the AI generated a structured prompt like this:

 "Create a 20-minute introductory talk on Artificial Intelligence for a non-technical college audience.
Include simple definitions, real-world applications like ChatGPT and facial recognition, 
and one visual example or demo.
 Emphasize practical relevance and ethical considerations." 

This method not only improves the prompt but also helps clarify the scope and flow of your final content.

Understanding the TCREI Prompt Framework 🔝

TCREI is a five-step method to write better AI prompts by focusing on clarity, purpose, and refinement:

• Task: Clearly define what the AI needs to do.
• Context: Provide background information like audience, tone, or usage scenario.
• References: Suggest a format, structure, or example the AI should follow.
• Evaluate: Check if the output meets your expectations for clarity, tone, and relevance.
• Iterate: Refine the prompt based on the results and repeat if needed.

🧪 Try Your Prompts in Google Colab 🔝

Below is a screenshot from my own Colab session showing the agent generating code and results live. It demonstrates how TCREI prompts translate into actionable insights.

Writing a great prompt is just the first step — seeing it in action is where the learning really happens. Google Colab now includes an AI-powered Data Science Agent that lets you test your prompts directly on uploaded datasets. Whether you're generating code, debugging, or analyzing CSV files, Colab offers an interactive environment to refine and evaluate your AI prompts in real time.

What is Colab ? Data Science Agent to get Insights from Stack Overflow Developer Survey

Try asking in Colab: “What are the top 5 programming languages used by survey respondents?” — then run the plan and explore results.

🔧 Tip: If the AI-generated code errors, paste both code & traceback into Colab and ask: “Explain this error and fix it” — the AI can guide debugging.

Get Hands-On! 🔝

Use the prompts and TCREI framework above. Then head to Colab’s Data Science Agent and test one in real time.

🔍 Real-World Prompt Examples 🔝

🔎 Advanced Concept: NLP-Friendly Content Prompts 🔝

An NLP-friendly prompt is crafted to work effectively with AI language models. It helps ensure outputs that are well-structured, semantically rich, and aligned with user intent and SEO goals.

✨ Key Traits of NLP-Friendly Prompts

• Clarity & Specificity: Leave no room for misinterpretation.
• Contextual Guidance: Define tone, audience, and purpose.
• Semantic Keywords: Use related terms to enrich understanding.
• User Intent: Match what your audience actually needs.
• Conversational Style: Natural language improves results.

Before: "Write about accounting software."
After: "Write a blog post for small business owners about the benefits of using cloud-based accounting software in 2025. Highlight real-time collaboration, expense tracking, and data security. Use a friendly, informative tone."

🎯 Final Thoughts 🔝

Writing great prompts is a learnable skill. Using the TCREI Prompt Framework gives you a strong foundation, and applying NLP-friendly techniques takes your prompting to the next level. The key is to experiment, adjust, and always be clear about your goal. The better your prompt, the better your AI.

👤 About the Author

Subhendu Mohapatra is the creator of Plus2net.com and a dedicated developer focused on AI-powered tools, data analysis, and content automation. He regularly experiments with platforms like Google Colab, Python data workflows, and prompt engineering to explore practical uses of AI in digital content and analytics.

Driven by a passion for knowledge sharing, he helps others build technical skills and leverage AI more effectively in their personal and professional workflows—often contributing on a voluntary basis through tutorials, code samples, and real-world guidance.

🎥 Join me live on YouTube

← Subscribe to our YouTube Channel here