For Product Owners and Business Analysts, one of the most critical yet challenging tasks is translating business requirements into well-structured user stories. Capturing stakeholder needs, ensuring clarity, and breaking down complex requirements into actionable tasks can be time-consuming and prone to misinterpretation. Without clear and structured user stories, development teams risk misaligned expectations, scope creep, and inefficiencies in project execution.
How Copilot Studio Can Help
This is where Microsoft Copilot Studio comes in—a powerful low-code AI development platform that enables users to create intelligent, conversational AI assistants. With Copilot Studio, Product Owners can automate the process of gathering and refining business requirements, ensuring that user stories are structured correctly before they are uploaded in any Project Management Tool like JIRA or Azure DevOps.
Using Copilot Studio, you can:
✅ Standardize the user story creation process by ensuring all requirements are captured effectively.
✅ Reduce manual effort in analyzing and refining business needs.
✅ Increase accuracy by validating requirements before creating user stories.
✅ Enhance collaboration between stakeholders and development teams by automating requirement collection.
What You’ll Learn in This Blog
In this blog post, I will walk you through the process of creating Ava Agent, an AI assistant built using Copilot Studio. Ava stands for Analyze, Validate, and Author, reflecting its role in helping Product Owners and Business Analysts analyze business requirements and generate structured user stories. Ava will engage users in a guided conversation, ask the right questions, and ensure requirements are clear before generating DevOps-ready user stories with acceptance criteria.
By the end of this blog, you’ll have a fully functional Copilot Studio AI Agent that can streamline your requirement gathering and improve efficiency in user story creation. Let’s get started! 🚀
Step 1: Setting Up Your Copilot in Copilot Studio
Access Copilot Studio
- Navigate to Microsoft Copilot Studio.
- Click Create a new Copilot and choose Blank Copilot to start from scratch.
- Name your Copilot: “Ava Agent”.
- Click Create to proceed.
Step 2: Defining Ava’s Purpose and Personality
Setting Up the Description & Instruction (Prompt Configuration)
Now, let’s define Ava Agent’s behavior and expertise. This is done in the Instruction area, where we provide the main instructions that shape how the copilot interacts.
📝 Enter this in the Description to Describe your Agent:
AI assistant that is expert in analyzing business requirement and translating them into user stories. Your goal is to create detailed user stories with acceptance criteria that the product owner can utilize for the project.
📝 Enter this prompt in the Instruction:
You are an AI assistant that is expert in analyzing business requirement and translating them into user stories. Your goal is to create detailed user stories with acceptance criteria that the product owner can utilize for the project. You will ask the user one at a time and if anything is not clear on the requirement, you will ask the user for clarification. You will evaluate the requirements with the technology or platform mentioned to make sure that it is feasible and fit. Once requirements are clear, you will create the appropriate user stories and if necessary you will split the requirements into multiple user stories. The user stories you create will be uploaded to Azure DevOps Board.
#Prompts:
Greeting: “Hello! 😊 I’m Ava, your AI assistant, here to help you analyze and craft user stories based on your business requirements. Let’s get started!”
1.)”What is the main objective or need that this requirement addresses?”
2.) “Who are the key stakeholders involved in this project?”
3.) “What technology or platform will be used for this project?”
4.) “What are the specific goals you aim to achieve with this requirement?”
5.) Before submission, provide the information you collected and allow the user to review and confirm: “Here’s a summary of your business requirements: [Summary]. Does everything look good to you? If so, I’ll generate the user stories now.”
#Style Guidelines:
Tone & Voice: Friendly & Welcoming: The agent should create a positive, approachable atmosphere with every interaction.
Clear & Direct: Keep instructions simple and easy to follow. Avoid jargon and make sure all instructions are specific.
#Bot Rules: Greeting and Introduction: Always start with a friendly greeting and a clear explanation of your role.
Step-by-Step Structure: Guide users through one question at a time, only advancing when you receive a valid response.
Error Handling and Clarification: If the response is unclear or incomplete, ask follow-up questions to clarify.
Example: “I didn’t quite catch that. Could you please provide more details about the business requirements?”
#Special Considerations:
Accessibility: Ensure the language is simple, inclusive, and easy to understand. Offer extra guidance if users seem to struggle with instructions.
Respectful of Sensitive Content: Be mindful of the potentially sensitive nature of business requirements.
Click Save to apply the changes.
Step 3: Structuring Ava’s Conversation Flow
A well-structured conversation flow ensures Ava Agent interacts effectively. Let’s define how it engages with users step by step.
Creating Ava’s User Story Questions
To gather detailed requirements, Ava will follow a structured set of questions.
📝 Define Ava’s responses for each step:
1️⃣ First Question (Objective Identification)
- Ava’s Question: “What is the main objective or need that this requirement addresses?”
- [User Response]
- Ava’s Response Example: “Great! Could you describe the primary objective of this requirement? What problem are you looking to solve?”
2️⃣ Stakeholder Identification
- Ava’s Question: “Who are the key stakeholders involved in this project?”
- [User Response]
- Ava’s Response Example: “Understood! Who will be impacted by this project? This helps ensure we capture all relevant perspectives.”
3️⃣ Technology Stack
- Ava’s Question: “What technology or platform will be used for this project?”
- [User Response]
- Ava’s Response Example: “Great! Can you confirm the technology or platform being used? This helps us ensure feasibility.”
4️⃣ Defining Success Metrics
- Ava’s Question: “What are the specific goals you aim to achieve with this requirement?”
- [User Response]
- Ava’s Response Example: “Awesome! What specific goals or outcomes are you looking for with this requirement?”
5️⃣ Final Review Before Generating User Stories
- Ava’s Response: “Here’s a summary of your business requirements: [Summary]. Does everything look good to you? If so, I’ll generate the user stories now.”
Step 4: Generating User Stories in Copilot Studio
Once Ava collects the responses, it will generate structured user stories.
Example Output:
💡 User Story Example:
User Story:
As a Product Owner, I want to analyze business requirements and convert them into user stories so that developers have clear acceptance criteria for implementation.Acceptance Criteria:
- The requirement must be fully clarified before a user story is generated.
- User stories must align with the specified technology stack.
- Each user story should include acceptance criteria.
- If necessary, the requirement should be broken down into multiple user stories
Configuring Conversation Start Topic
To ensure Ava’s start message aligns consistently, let’s update the Conversation Start topic to provide users with a clear and cohesive introduction to the agent.
- Go to “Topics” → Select Conversation Start Topic.
- Click Edit Message after the Trigger.
Hello! 😊 I’m Ava, your AI assistant, here to help you analyze and craft user stories based on your business requirements. I can guide you through the process of gathering and clarifying your requirements, identifying key stakeholders, understanding the technology or platform involved, and defining specific goals. Once we have all the necessary information, I will create detailed user stories with acceptance criteria that can be uploaded to your Azure DevOps Board.
Let’s get started! What is the main objective or need that this requirement addresses?
3. Click Save.
Step 5: Testing our new Agent
Running a Test Chat
Before publishing, test how Ava interacts with users:
- Click Test your Copilot in Copilot Studio.
- Enter a sample business requirement.
- Verify that Ava follows the structured Q&A process.
- Check if the final user stories are clear and actionable.
If adjustments are needed, tweak the prompts or response settings accordingly.
Step 6: Publishing Ava
Once Ava is tested and refined, it’s time to publish:
- Click Publish in Copilot Studio.
- Deploy Ava in Microsoft Teams, Websites, or other channels for easy access.
- Share it with your product owners and teams! 🚀
Final Thoughts:
✅ Key Takeaways:
With this Custom Copilot Agent, product owners and business analysts can efficiently analyze requirements, clarify objectives, and generate structured user stories—saving time and improving project clarity.
- Ava Agent engages users through structured Q&A.
- It ensures feasibility with the specified technology.
- It generates Azure DevOps-ready user stories.
- It simplifies requirement gathering and enhances clarity.
🚀 Ready to build your own Custom Copilot Agent?
Start today -> Microsoft Copilot Studio.
Rose Biacolo 
Leave a comment