Effective Prompting
Learn how to write effective prompts using the 4 building blocks framework. Better prompts lead to better results with AI tools.
Core Skill
Effective prompting
Prompting is the core skill for working effectively with AI. The difference between a weak and strong prompt determines whether you get an unusable answer or output you can use immediately. Those who learn to prompt well consistently get more value from AI tools and save time by needing less rewriting and adjusting.
Downloads
What is prompting?
Prompting is giving instructions to an AI language model. A prompt is the input you provide: the question, task, or context that directly determines the quality of the output you receive.
Think of prompting as briefing a colleague. The clearer and more specific your brief, the better the result. A vague question leads to a vague answer. A structured, context-rich prompt delivers relevant, usable content.
Prompting is the core skill for working effectively with AI. The difference between a weak and a strong prompt can mean the difference between a useless answer and output you can use immediately.
AI models are prediction machines
AI predicts the most likely next word based on your input and patterns from billions of texts. It doesn't truly understand what you mean: it generates text based on probabilities. That's why a good prompt is crucial: it guides the model in the right direction.
The 4 Building Blocks of a strong prompt
Use the 4 Building Blocks Framework to systematically write better prompts. You don't always need all four building blocks, but they help you structure what you want to achieve.
1. TASK – What should the AI do?
Give a clear instruction. Be specific about what you want: write, summarize, rewrite, translate, brainstorm, or analyze.
Examples:
- "Write an email to..."
- "Summarize this text in 3 bullet points"
- "Generate 5 headlines for..."
- "Analyze this feedback and identify patterns"
2. CONTEXT – What should the AI know?
Provide relevant background information. This is often the difference between a generic and a useful answer. Include: target audience, tone of voice, brand guidelines, relevant data, constraints.
Examples:
- "This email is for an existing customer who's been with us for 3 years"
- "Write in this tone of voice: accessible, personal, no-nonsense"
- "Target audience: marketers aged 30-45"
- "Context: We're launching this product in Q1 2026"
3. ROLE – Who should the AI be?
Give the model a role or perspective. This influences the tone of voice, word choice, and depth of the answer.
Examples:
- "You are a senior content marketer"
- "You are a customer service representative with 10 years of experience"
- "You are a copywriter specializing in conversion optimization"
- "You are an SEO expert who understands how search engines work"
4. FORMAT – What should the output look like?
Specify the desired structure and length. This prevents getting an essay when you wanted bullet points, or vice versa.
Examples:
- "Give the answer in 3 bullet points, maximum 50 words per point"
- "Write an email of maximum 150 words"
- "Create a table with columns: Name, Type, Price, Status"
- "Use a friendly tone of voice and markdown formatting"
When do you use which building blocks?
You don't always need all four building blocks. For a simple question, TASK is often enough. For more complex assignments, CONTEXT, ROLE, and FORMAT are crucial for a usable result.
Rule of thumb: The more you want to use the output directly, the more building blocks you need.
Practical prompt tips
Besides the 4 building blocks, there are general principles that make your prompts stronger. These tips come from training and best practices of experienced AI users.
Ask questions first
Let AI ask questions before executing a task. Add: "What questions do you have before we start?"
Do it step by step
For complex tasks: ask AI to work step by step. This activates "chain-of-thought reasoning" for better answers.
Specify your audience
AI is trained primarily on English data. Explicitly specify the target audience and cultural context for your output.
Be specific and concrete
Vague prompts lead to vague answers. Specify what "better" means instead of just "Make it better".
Iterate and refine
See your first prompt as a starting point. Refine based on the output: "Make this shorter", "Add more examples". AI models build on conversation history, so working iteratively is more efficient than starting over each time.
Practical examples
Below you'll see the difference between weak and strong prompts. Notice how the 4 building blocks and prompt tips are applied.
Example 1: Writing an email
X Zwakke prompt:
✓ Sterke prompt:
Applied building blocks:
- TASK: Write an email
- CONTEXT: Existing customers 2+ years, new AI search functionality
- ROLE: Product marketer
- FORMAT: Maximum 150 words, enthusiastic but professional, call-to-action
Example 2: Content brainstorming
X Zwakke prompt:
✓ Sterke prompt:
Applied building blocks:
- TASK: Generate 5 LinkedIn post ideas
- CONTEXT: B2B, target audience marketers age 30-50
- ROLE: Social media strategist specializing in B2B content
- FORMAT: Each idea with opening line, core message, visual element
- Tip: "What questions do you have before we start?"
Example 3: Summarizing text
X Zwakke prompt:
✓ Sterke prompt:
Applied building blocks:
- TASK: Summarize 5 pages of market research
- CONTEXT: Focus on findings, opportunities, and risks
- FORMAT: 3 bullet points, max 40 words per point
- Tip: "Do it step by step"
From weak to strong: The pattern
Weak prompts are vague and generic. Strong prompts are specific and structured. The difference? Systematically applying the 4 building blocks and integrating prompt tips.
Start with a weak prompt if you need to, but use the output to iterate toward a strong prompt that's reusable.