📚 Prompt Typology & Use Cases

0 of 5 lessons • 0% complete

Lesson 1 of 5

The 8 Essential Prompt Types

📖 5 min✨ 30 XP

Not all prompts are created equal. Different tasks require different prompt structures. Understanding the 8 core prompt types helps you choose the right tool for each job.

1. Instructional Prompts

Direct commands telling the AI exactly what to do. Best for straightforward, single-step tasks.

Example: "Summarize this article in 3 bullet points."
Example: "Translate this email to Spanish."
Example: "Extract all email addresses from this text."

When to Use

Use instructional prompts when the task is clear, simple, and doesn't require context or creativity.

2. Role-Based Prompts

Ask the AI to adopt a specific persona or expertise. Taps into the AI's training on that role.

Example: "Act as an experienced financial advisor. Review this investment portfolio and suggest rebalancing strategies."
Example: "You are a skeptical journalist. Identify potential weaknesses in this press release."

Power Tip

Role-based prompts work because AI models are trained on vast amounts of role-specific text. Saying "act as a lawyer" activates patterns from legal documents.

3. Constrained Output Prompts

Heavy focus on format and structure. Best for data extraction, API responses, or strict formatting needs.

Example: "Return ONLY valid JSON with keys: name, email, phone. No additional text."
Example: "List exactly 5 items, numbered, with no explanations."

4. Few-Shot Prompts

Provide examples for the AI to learn from. Pattern-based learning.

Example:
"Classify these customer reviews as Positive, Neutral, or Negative:

Review: 'Amazing product, works perfectly!' → Positive
Review: 'It's okay, nothing special.' → Neutral
Review: 'Completely broken, waste of money.' → Negative

Now classify: 'Best purchase I made this year!'"

5. Chain-of-Thought Prompts

Explicitly ask the AI to show its reasoning step-by-step. Dramatically improves accuracy on complex reasoning tasks.

Example: "Solve this problem step-by-step, showing your reasoning:

If a train leaves Chicago at 60 mph heading west, and another leaves Denver at 80 mph heading east, when will they meet? Chicago to Denver is 1,000 miles.

Think through this step by step."

Magic Phrase

Adding "Let's think step by step" or "Show your reasoning" can improve accuracy by 20-30% on math and logic tasks.

6. Conversational Prompts

Multi-turn dialogue where each response builds on the previous. Best for exploratory tasks.

Prompt 1: "What are the main causes of customer churn in SaaS?"
Prompt 2: "Focus on the onboarding phase. What specific issues happen there?"
Prompt 3: "Design a 3-step solution to fix onboarding churn."

7. Creative Prompts

Open-ended prompts for brainstorming, ideation, or content creation. Less constraints, more freedom.

Example: "Generate 10 creative tagline ideas for an eco-friendly water bottle brand targeting millennials."
Example: "Write a compelling opening paragraph for a mystery novel set in 1920s Paris."

8. Zero-Shot Prompts

No examples, no context—just the task. Relies entirely on the AI's training. Works for common, well-understood tasks.

Example: "What is photosynthesis?"
Example: "Define machine learning."

Lesson 1 / 5