When you interact with AI models, the quality of the response depends largely on how well you frame your request. If you’re not getting the best responses, it’s not necessarily the AI’s fault — it might be time to refine your prompt. This process, known as “prompt engineering,” is about asking questions and giving instructions to make it easier for the AI to understand exactly what you want.

In this post, we’ll explore how better prompts always lead to better results, with easy-to-understand examples on topics like effective emails, travel planning, and more.

What is a Prompt?

A prompt is what you input into an AI model — a question, instruction, or request. It can be as simple as, “What’s the weather like today?” or more complex like, “Help me create a one-week travel itinerary in Europe.” The clearer and more detailed your prompt is, the better the AI can respond.

Example 1: Writing Effective Emails

Let’s say you need to write an important email and decide to ask an AI for help. Whether you’re sending an apology email, a business pitch, or a travel delay notification, the quality of the email depends on how you phrase the request.

Not-So-Great Prompt: “Write me an email.”

This is too vague. The AI doesn’t know what kind of email you need — personal, professional, formal, or casual?

Better Prompt: “Write me a professional email apologizing to a client for a two-day project delay. Mention that the delay is due to an unexpected issue and explain that the extra time will allow for better quality. Keep the tone formal yet empathetic.”

In this prompt, you provide the AI with specific details: the email’s purpose, the tone, and important context (the project delay and reason). As a result, the AI will generate a much more polished and relevant email for your needs.

Another Example: You need to follow up on a job application.

Not-So-Great Prompt: “Write a follow-up email about my job application.”

Better Prompt: “Write a follow-up email to a potential employer about my application for the software engineer position submitted two weeks ago. Ask politely for an update and express continued interest in the role. Keep it professional and concise.”

With a clear and specific prompt, the AI will produce an email that covers all the important details and uses the right tone.

Example 2: Planning a Travel Itinerary

Planning a trip is another situation where prompt engineering can make a big difference. Let’s say you want help organizing a travel itinerary.

Not-So-Great Prompt: “Make a travel plan for me.”

The AI might return a general suggestion, like visiting popular landmarks, but it won’t fit your specific needs unless you provide more information.

Better Prompt: “Create a 7-day travel itinerary for a family vacation in Japan. Include visits to cultural sites in Tokyo and Kyoto, outdoor activities like hiking or hot springs, and recommend family-friendly restaurants. We prefer a relaxed pace and want to avoid tourist traps.”

This prompt gives the AI clear direction on the length of the trip, the destinations, activities, and preferences. You’ll get an itinerary that’s much closer to what you envision.

Another Example: You want to plan a budget-friendly trip to Europe.

Not-So-Great Prompt: “Plan a trip to Europe for me.”

Better Prompt: “Plan a budget-friendly, 10-day solo trip to Europe with stops in Paris, Amsterdam, and Berlin. I want to spend no more than $150 per day, including accommodation, meals, and transportation. Recommend affordable hostels, cheap eats, and public transportation options.”

By including your budget, travel style (solo), and preferred cities, the AI can tailor the plan to fit your financial and personal preferences.

Example 3: The Power of Specificity

Let’s revisit the idea of asking for a recipe.

Not-So-Great Prompt: “Give me a recipe.”

You’ll get a random dish that might not suit your taste, dietary restrictions, or time constraints.

Better Prompt: “Give me a vegetarian dinner recipe that takes less than 30 minutes to prepare and uses ingredients I can find at a typical grocery store.”

Now the AI understands exactly what you want: vegetarian, quick, and easy-to-find ingredients. The output will be far more relevant and useful to you.

Example 4: Breaking Complex Tasks into Steps

Sometimes, a question is too broad, and the AI will struggle to give a helpful answer.

For example, you’re looking for career advice.

Not-So-Great Prompt: “How can I improve my career?”

This is also vague. The AI might offer generic advice that isn’t personalized to your field.

Better Prompt: “What skills should I develop if I want to advance in my software engineering career over the next 5 years? Please suggest both technical and soft skills.”

By narrowing down your question to specific goals, the AI can offer targeted suggestions that will actually help you.

Example 5: Developing a Project Idea

Imagine you have a rough idea for a new project but need the AI to help flesh it out.

Not-So-Great Prompt: “Give me a project idea.”

This prompt is too vague and will result in a random or generic project idea, which may not fit your interests or field.

Better Prompt: “I’m interested in building a mobile app that helps people track their daily habits and progress. Suggest a project idea that includes unique features like reminders, goal-setting, and data visualization. Also, suggest technologies or frameworks I could use to develop it.”

In this prompt, you give clear information about your project focus (habit tracking), and features you’re considering, and even request recommendations on technologies. The AI will now provide a more well-rounded and specific project idea, including potential implementation strategies.

Another Example: Suppose you’re seeking feedback on an existing project idea.

Not-So-Great Prompt: “Analyze my project idea.”

Without context, the AI has little to go on, and the analysis might be too general to be helpful.

Better Prompt: “Analyze my idea for a web-based marketplace that connects local artists with buyers. The platform should support user profiles, secure payments, and product reviews. Please identify potential challenges with scalability, user adoption, and competition. Offer suggestions on how to overcome these issues.”

Here, you’ve provided a detailed project description and specific areas you’d like feedback on (scalability, user adoption, and competition). The AI can now deliver a thoughtful analysis with actionable insights.

Example 6: Creating a Presentation

Creating an engaging presentation can be challenging, but AI can help — if you provide the right prompt.

Not-So-Great Prompt: “Create a presentation for me.”

This is too open-ended and will lead to a generic response that likely won’t meet your needs.

Better Prompt: “Create a 10-slide presentation on the importance of cybersecurity in small businesses. The presentation should cover key threats, simple security measures, and examples of real-life cyberattacks on small companies. Please include a slide with statistics and a concluding slide with action steps.”

By giving specific information about the presentation’s topic, structure, and key content areas, you ensure the AI produces a more relevant and organized set of slides.

Another Example: You want a visually appealing presentation for a marketing pitch.

Not-So-Great Prompt: “Help me with a marketing pitch presentation.”

Without knowing your audience, product, or the format, the AI may create something too broad.

Better Prompt: “Create a 15-slide presentation for a marketing pitch about our new organic skincare line. The target audience is eco-conscious consumers, and we want to highlight the product’s natural ingredients, sustainability, and customer testimonials. Include slides for a product demo, pricing, and a closing call to action.”

Here, you provide details on the product, audience, and structure, which ensures the AI delivers a highly relevant and tailored presentation.

Conclusion: Practice Makes Perfect

Better prompts always lead to better results. The key is to be clear, specific, and, when necessary, provide context or examples. You don’t need to be an expert to start using these strategies; with a little practice, you’ll see a big difference in the quality of your AI-generated content.

Whether you’re writing emails, planning a vacation, analyzing a project, or creating a presentation, the more details and context you include, the better the AI will understand your needs. So spend a little extra time refining your prompt — it’s worth it!

Happy prompting!