Lesson 3.3: Common Prompt Patterns
Introduction
To use ChatGPT effectively in AI automation, understanding common prompt patterns is essential. Prompt patterns are structured ways of instructing AI to produce consistent and reliable outputs. This lesson introduces the most widely used prompt patterns and explains how they can be applied in automation workflows.
Pattern 1: Instruction-Based Prompts
Instruction-based prompts clearly tell the AI what to do. They are simple and direct. Examples:
-
“Summarize this text in three bullet points.”
-
“Translate the following paragraph into English.”
-
“Generate a list of email subject lines for a marketing campaign.”
These prompts work well for tasks that have a specific and predictable outcome.
Pattern 2: Role-Based Prompts
Role-based prompts ask the AI to adopt a specific persona or function. This improves output relevance. Examples:
-
“You are a professional editor. Improve the following text.”
-
“Act as a content strategist and suggest blog topics for beginners.”
-
“You are an assistant. Draft a polite reply to this customer email.”
Role-based prompts help AI understand the tone, style, and perspective required.
Pattern 3: Step-by-Step or Multi-Step Prompts
Some tasks require multiple steps. In such cases, breaking instructions into steps ensures accuracy. Examples:
-
“First, summarize the report. Then, create a bullet-point list of key actions.”
-
“Analyze the text, extract important keywords, and then generate a short summary.”
Step-by-step prompts reduce errors and increase the quality of complex outputs.
Pattern 4: Example-Based Prompts
Providing an example of the desired output guides the AI and improves consistency. Examples:
-
“Here is an example of a weekly email draft: [sample]. Create a similar draft for this week.”
-
“Use the table format shown below to organize this data.”
This technique is particularly useful when output structure or style is important.
Reflection Exercise
Think of a repetitive task in your workflow. Identify which prompt pattern—Instruction-Based, Role-Based, Step-by-Step, or Example-Based—would be most effective. Write a sample prompt using that pattern.
Key Takeaways
-
Common prompt patterns improve consistency and accuracy in AI outputs
-
Instruction-Based prompts are direct and simple
-
Role-Based prompts guide tone and perspective
-
Step-by-Step prompts break complex tasks into manageable steps
-
Example-Based prompts provide clear output structure for better results
