- 18 Dec 2024
- 1 Minute to read
- Print
- DarkLight
- PDF
Creating a PRD
- Updated on 18 Dec 2024
- 1 Minute to read
- Print
- DarkLight
- PDF
In the PRD options window, you can choose from multiple methods to create a PRD, depending on your situation and goals.
Some options are available only on specific configurations. If an option is not available in your system, contact our Support team.
How Would You Like to Start Building Your PRD?
Create a PRD from Scratch
Answer a few simple questions, and PRD-AI generates a custom PRD that includes:
A structured list of feature requirements
Non-functional requirements
Acceptance criteria
This is ideal when you’re starting fresh and want a guided, AI-assisted approach.
Enhance Your Existing PRD
Upload your current PRD to Crowdbotics and continue refining and improving it. PRD-AI helps you:
Identify gaps or incomplete sections
Suggest additional features or enhancements
Improve overall clarity and structure
This method is perfect for scaling or improving an existing product.
Create an Empty PRD
Start with a blank slate and fully customize your PRD to meet your exact requirements. You have complete control to:
Add features, requirements, and acceptance criteria manually
Use PRD-AI to generate recommendations as needed
This is great for teams who already know what they want but prefer to build the PRD step by step.
Generate a PRD from a Document
Upload a technical document, schematic, or requirements file, and PRD-AI transforms it into a structured PRD. The platform:
Analyzes your document’s content
Extracts features, requirements, and components
Builds a comprehensive PRD based on the input
This is ideal if you already have detailed technical documentation that needs structuring.
Create a PRD from a GitHub Repository
Select a GitHub repository, and PRD-AI analyzes your codebase to generate a PRD.
This process:
Provides insights into the code structure, functionality, and key components
Summarizes existing features and dependencies
Builds a PRD with all identified requirements
Best suited for modernizing existing codebases or understanding complex, legacy projects.