The difference between an AI prompt that works and a prompt that works reliably is structure. The 4P Framework is a prompt engineering method that delivers more consistent, predictable, and better results.

 

You'll learn to craft prompts using four essential elements: ​

 

• Persona: the role AI assumes

• Purpose: the task and its goals

• Process: the steps AI should follow

• Pattern: the format of the output

​

With this framework, you'll be able to design good prompts faster, troubleshoot underperforming prompts, and build a library of reliable go-to prompts you can adapt to many projects.
 

Learning Objectives: 

• Identify the four components of the 4P Framework and their function in a prompt

• Construct prompts using Persona, Purpose, Process, and Pattern

• Apply the 4P Framework to common editorial tasks

AI edits shouldn't happen in a black box. When you can't see why the AI made a change, you lose control—and control is what separates editorial judgment from automated output. In this lesson, you'll learn prompting techniques that force AI to show its work:
​

• Propose Alternatives: Get multiple revision options with the reasoning behind each, giving you choices instead of a single take-it-or-leave-it output.

• Confidence Signals: Ask AI to rate its certainty, helping you prioritize focus your review.

• Disagreement Flags: Prompt AI to flag edits where reasonable editors might differ, surfacing subjectivity so you can make the final call.

Then we'll extend these transparency techniques to rubric prompting, where we’ll have AI create scoring criteria, build a rubric, and assess content against it, with scores and reasoning clearly displayed so you can turn vague AI judgment into documented, repeatable rationale.​

​​

Learning Objectives:

• Apply "show your work" techniques to maintain editorial control over AI outputs

• Construct prompts that produce transparent, auditable AI reasoning

• Build and implement rubrics to evaluate content and compare options against defined criteria

In AI editing, the best results often don't come from telling it longer prompts or more detailed descriptions. They come from showing the AI what you want and guiding it through the process.

First, we'll cover how to use examples in your prompts. AI learns better when you show it what you're after, just like a picture being worth a thousand words. This technique is especially powerful when you're matching a specific voice or brand or trying to capture a "feel" that's hard to put into words. You'll learn:

 

• Format to Follow: providing a sample structure for AI to match

• Like/Don't Like: showing both a good and bad example to steer the AI

Then we'll explore Chain of Thought prompting, a method for breaking complex tasks into smaller steps. Instead of asking AI to do everything at once, you'll list subtasks and prompt AI to work through them one by one. This approach leads to better results because the AI gives each step more attention, and you can review, refine, or redirect along the way.

​

Learning Objectives:

• Apply Few Shot prompting to achieve consistent, on-target outputs

• Construct Chain of Thought prompts that break complex tasks into manageable steps

• Combine these techniques to improve quality and maintain control over multi-step editorial tasks
  ​​​

When AI misses the mark, the issue often isn't what you asked for, it's that the AI struggled to process the instructions. In this lesson, you'll learn two techniques that give the AI a clearer roadmap to success: Markdown and XML.

 

Markdown is a simple way to add formatting—bold, italics, headings, lists—using basic symbols. It makes your formatting portable, so structure survives the copy/paste from AI into Word, Google Docs, or wherever your text needs to go. 

 

XML tags are a labeling system that helps AI understand the meaning and structure of information. They let you clearly separate the task, content, and expected output so the AI processes each part the way you intend.

 

Used together, they give you precise control over both the AI's behavior and the finished output. You don't need to learn to code to use these techniques—we'll cover how to get AI to add this structure for you. We'll also combine these methods to build custom style guides that help AI reliably edit text according to your organization's rules.

​

Learning Objectives:

• Apply Markdown formatting to create portable, consistently structured outputs

• Use XML tags to organize complex prompts and knowledge documents

• Combine Markdown and XML tags to make a custom style guide AI can read

 

Prompts tell AI what to do in the moment. Knowledge documents tell the AI what to know every time. In this lesson, you'll learn how to build reference documents that work alongside your prompts so the AI can apply your standards consistently without you repeating yourself.

We'll focus on three types of knowledge documents:

​

• Custom Style Guide: your house rules for grammar, punctuation, formatting, and usage

• Content Examples: samples that show AI the tone, voice, or format to follow

• Restricted Terms: terms to skip for brand, legal, or inclusivity reasons, along with preferred alternatives

 

This is how you scale your editorial standards without sacrificing control. Build the documents once, use them everywhere, and have the AI apply your rules the same way every time.

​

Learning Objectives

• Identify which knowledge document types best support different editorial tasks

• Build knowledge documents using Markdown, XML, and spreadsheet formats

• Combine knowledge documents with prompts for consistent, standards-aligned AI output

​​