Prompt Engineering Tips
The more specific your prompt, the better the AI can understand what you're looking for. Vague prompts lead to vague responses.
❌ Weak Prompt:
"Tell me about cars."
✅ Strong Prompt:
"Explain the key differences between electric and hybrid cars, focusing on environmental impact, cost of ownership, and performance."
Providing context about who you are, what you're trying to accomplish, and your level of familiarity with the topic helps the AI tailor its response to your needs.
❌ Weak Prompt:
"How do I code a website?"
✅ Strong Prompt:
"I'm a marketing professional with no coding experience. I need to create a simple portfolio website to showcase my work. What are the easiest tools or platforms I could use, and what basic steps should I follow?"
The RICCE framework helps you create comprehensive prompts that get better results:
- Role: Specify who the AI should act as (e.g., "As a financial advisor...")
- Instruction: Clearly state what you want the AI to do
- Context: Provide relevant background information
- Criteria: Define what makes a good response
- Example: Show an example of what you're looking for (optional)
✅ RICCE Example:
"Role: As a social media marketing expert
Instruction: Create a content calendar for a new coffee shop
Context: The shop is opening next month in a college town, targeting students and young professionals
Criteria: Include 3 posts per week across Instagram and TikTok, with a mix of promotional content and coffee education
Example: Similar to how Blue Bottle Coffee structures their social media presence"
Don't expect perfect results from your first prompt. Use the AI's response to refine your next prompt, gradually steering toward the outcome you want.
If the response isn't what you expected, try:
- Asking the AI to expand on specific parts
- Requesting a different approach or perspective
- Providing feedback on what was helpful and what wasn't
- Adding constraints or additional criteria