Developing Agile Requirements with AI

An effective agile project starts with sound requirements gathering. Integrating AI can also make it more productive. Let’s look at approaches you can take to collect requirements with assistance from AI.

First, you can use AI to give you a set of initial requirements by defining the purpose of the agile project. (A simple prompt is at the end of this article.) From there, you can add the particulars of your project and environment to get a “starter kit” for requirements gathering. Distribute this and have your stakeholders share additions and clarifications to the requirements starter kit. It will make everyone more productive.

The feedback you receive regarding your starter kit requirements will undoubtedly generate new ideas. Encourage your stakeholders to concentrate their ideas on solving existing business problems or leveraging improvement opportunities. Their responses can be in the form of Word documents, emails, or you can record and transcribe conversations.

And here’s where AI can jump-start your productivity.

Copying your stakeholder feedback into a GPT tool can help you accelerate many activities. These include:

• Applying Natural Language Programming (NLP) to rapidly analyze the feedback. This can detect contradictions and identify points that need to be clarified. This sorts out the next steps you need to take to create sound requirements.
• Create a first pass at prioritizing features. After reviewing the output, it can be distributed to key stakeholders for validation and used during the first sprint.
• Help make your requirements into “executable statements” so the next stage – writing code – is more efficient and accurate.

Remember, there are drawbacks to keep in mind when using AI tools.

• First, remember it is an agile project you are preparing. Don’t get too crazy creating detailed requirements by running too many AI tool iterations! The agile approach itself will refine the requirements in a much more focused manner.
• Second, always be mindful of the data copied into an AI tool. It is easy to raise privacy concerns, even if the organization has a protected AI environment. Be aware not to compromise the company’s confidential information.
• Finally, any output from an AI tool can contain errors or incorrect inferences, typically called hallucinations. Review AI products before distribution.

++++

Sample prompt – This can provide a set of requirements to discuss with stakeholders, and be expanded with follow-up queries with specific information from your project.

Prompt:  As an agile scrum master with at least 10 years of experience, provide an example of 20 requirements for an agile project for developing a dashboard for executives to use to obtain information about the status of their business quickly, and more easily make strategic decisions.

RESPONSE: To develop a dashboard for executives in an agile project, you can consider the following 20 requirements:

1. User-friendly interface for quick access to information.
2. Customizable KPIs and metrics based on business needs.
3. Real-time data integration for up-to-date information.
4. Interactive visualizations for easy data interpretation.
5. Mobile responsiveness for access on the go.
6. Integration with existing business systems and data sources.
7. Role-based access control for data security.
8. Support for drill-down capabilities to explore detailed information.
9. Predictive analytics for forecasting and trend analysis.
10. Data export functionality for further analysis.
11. Performance monitoring of critical business processes.
12. Alerts and notifications for important changes or thresholds.
13. Data quality management to ensure accuracy and reliability.
14. Support for multiple data formats and types.
15. Collaboration features for sharing insights and decisions.
16. Integration with task management and workflow tools.
17. Compliance with data privacy and governance regulations.
18. Executive dashboard templates for common use cases.
19. Feedback mechanisms for continuous improvement.
20. Regular iteration and enhancement based on user feedback and changing business needs.

These requirements are based on the principles of agile project management and the specific needs of developing a dashboard for executives to make strategic decisions.

Additional sample prompts are available as part of the LinkedIn Learning course titled AI-Powered Agile: Strategies for Modern Project Managers.