Module: Course & Education

Prompting for Non-Technical Users: Get More Out of AI Without Code

Learn to communicate effectively with artificial intelligence in everyday situations, without needing to be a programmer.

Introduction: What actually is a prompt?

You have probably heard of ChatGPT or other AI assistants. Perhaps you have already tried them, but the result was disappointing. Often, this is not due to the intelligence of the model, but rather the way we humans formulate our questions. Simply put, a prompt is the instruction or question you give to an AI.

Many users treat AI as if it were a traditional search engine (like Google): they type in loose keywords and hope to magically find the perfect answer right away. But AI works fundamentally differently: it is a digital sparring partner that responds to language, context, and roles. In this module, you will learn how to get the most out of AI in a few simple steps, completely without technical prior knowledge or code.

Key Message

AI does not have a built-in mind reader. The clearer, more detailed, and more logical your instruction is, the more useful the response will be.

The 4 Basic Principles of Effective Prompting

To get a good result, it helps to maintain a consistent structure. You can think of this as instructing a smart, new colleague who has never worked with you before.

1. Asking Clear Questions (Who, What, Why)

Be specific about what you are looking for. Do not ask for "something about marketing", but state exactly what you want to achieve.

Weak prompt

"Write an email for my customers."

Strong prompt

"Write a friendly email of up to 150 words to our regular customers to announce that our opening hours will change to 9:00 AM to 5:00 PM as of August 1st."

2. Providing Context

Provide background information. Who is your target audience? What is the tone? What background does the reader have? The more relevant context you share, the better the AI can put itself in the situation.

3. Iterating (Engaging in Dialogue)

Do not expect the answer to be one hundred percent perfect right away. Think of it as a conversation. If the tone is too formal, you say: "This is good, but make it a bit more casual and friendly, as if you are talking to a good acquaintance." By asking follow-up questions (iterating), you shape the response to your liking.

4. Managing Expectations

AI is a powerful tool, but it is not an oracle. It can make mistakes or make up facts (hallucinate). Always critically check the output, especially when it comes to numbers, legal matters, or important decisions.

Everyday Example: The Planning Email

Imagine: you are organizing a neighborhood gathering in your hometown and you need to send an invitation to your neighbors. You can tackle this directly with a well-thought-out prompt.

Practical prompt example

"You are an engaged neighbor. Write an invitation for an informal neighborhood get-together on Saturday afternoon at 4:00 PM in the street. The tone should be warm, welcoming, and accessible. Ask them to bring something tasty to share and to let us know by Wednesday at the latest if they are coming."

The result of such a prompt immediately delivers a ready-to-use draft text that you can send directly in the neighborhood app or via email after a minor adjustment.

Exercise: Write Your Own First Goal-Oriented Prompt

Time to try it yourself. Choose an everyday task from your own work or personal life (for example: drafting a friendly reminder for an outstanding invoice, planning a weekly menu for the family, or preparing for a job interview).

Formulate your prompt using this standard checklist:

  1. Role: Who should the AI be? (E.g., "Act as an experienced communications advisor...")
  2. Task: What exactly needs to happen? (E.g., "...and write a short reminder...")
  3. Context: What are the constraints? (E.g., "...for a client who is two weeks overdue, keep the tone friendly but clear.")
  4. Format: What should it look like? (E.g., "Provide a subject line and the email body.")

Test this prompt in your favorite AI tool and adjust the result via a follow-up question (iteration) if the tone is not yet completely to your liking.