It’s common to prompt an AI with a basic question or sentence such as: “Write me an article about supplements for bone support.”
While sometimes able to produce a good result, a prompt like this lacks specificity. In this example, it doesn’t explain the purpose of the content, the audience, the length, the tone of voice, whether to SEO optimize the article, or any other supplemental information that might improve the output.
Megaprompts are used with AIs like ChatGPT, GPT-4, GPT-3, Claude, and behind the scenes in AI tools like Jasper.ai, Copy.ai, and Writesonic.
Elements of a megaprompt
Not all megaprompts have all the below elements, but each one potentially offers an improvement in the output:
- Persona: What the AI should act as
- Task: What the AI should do
- Steps: The steps to complete the task
- Context: Context around the task, such as audience, length, level of detail, writing style, type of analysis and so forth
- Examples: Examples of inputs and outputs
- Goal: The desired results of the task if not specified elsewhere
- Format: How I should return the output, such as in a code block, markdown text, or bullet point list
TASK: OUTLINE THE FOLLOWING BOOK
ACT AS: NYT best-selling non-fiction author
GENRE: Business motivation, leadership, self-improvement
VOICE AND STYLE: clear, engaging, data-driven, support insights with examples and research, conversational tone to make complex ideas easy to understand, figurative, challenge common beliefs, encourage readers to question assumptions
SUSTAIN CURIOSITY WITH: 1) Counter-intuitive, 2) Counter-narrative, 3) Shock and awe, 4) Elegant articulations
WRITING TECHNIQUE: 1) use vivid words, 2) engage imagination, 3) occasionally use humor, 4) replace plain phrases with vivid metaphors, 5) remove unimportant details
TITLE: The Emotional Intelligence of Successful Entrepreneurs: How Self-Awareness and Empathy Drive Business Growth
OUTPUT FORMAT: Concise bullet points, indented bullet points, markdown
Problems with mega prompts
There are three main potential pitfalls of using megaprompts.
- Token count: Because megaprompts are longer than normal prompts, they consume more of the AI’s memory which is counted in tokens rather than words or characters. For AIs with smaller memories, this can pose problems as it leaves less room for the AI to interpret documents or write its output
- Complexity: Prompts that provide too many instructions in a single step can confuse the AI, creating too many associations and ideas for it to handle at the same time. As a result, overly complex prompts can produce lackluster results versus simpler instructions
- Lack of sequencing: While providing a set of steps has been shown to improve AI outputs, providing all of the direction in a single prompt is still asking the AI to do a lot in a single go. Often a better result can be achieved by breaking the steps down into multiple prompts and submitting them one-by-one to build on each other.