Topic 5: 7-Step approach to turn requirements into clear user stories
Generative AI can help Business Analysts turn raw requirements, workshop notes and stakeholder inputs into draft user stories. But simply asking AI to “create user stories” can produce incomplete requirements, hidden assumptions, or stories that look convincing without proper support.
The 7-step approach gives you a more structured way to work with GenAI so you can create clearer drafts while also identifying gaps, ambiguities, and unanswered questions that still need human attention.
How to apply the 7-step approach
1. Define the objective: Tell GenAI exactly what you are trying to achieve, for example, analyse the provided requirements and create draft user stories while identifying gaps, ambiguities and unanswered questions.
2. Provide relevant context: Give the AI the information it needs to understand the situation. This might include the business objective, workshop notes, stakeholder requirements, user or persona needs, the existing process or journey, business rules and system constraints.
3. Set constraints: Establish clear boundaries. Tell GenAI to work only with the information provided, avoid inventing missing requirements, clearly flag assumptions, distinguish confirmed requirements from gaps, and avoid resolving ambiguities without evidence.
4. Define the output: Specify exactly how you want the analysis structured. For example, ask for a table containing User story, Acceptance criteria, Business rules, Assumptions, Gaps/Questions and Source/Evidence.
5. Use examples when helpful: If your organisation has an approved user story format or specific terminology, provide an example. This helps GenAI understand the structure and level of detail you expect.
6. Review & challenge: Treat the generated stories as drafts, not finished requirements. Ask which requirements are ambiguous, what information is missing, which assumptions need validation, whether requirements conflict and which stories lack sufficient evidence.
7. Refine iteratively: Add stakeholder answers, clarify requirements, correct assumptions, resolve identified gaps and regenerate the affected stories. Continue until the output accurately reflects what is known.
What good GenAI-assisted BA work looks like
The goal isn’t simply to generate user stories faster. It is to use GenAI to help you structure what you know and surface what you don’t. A useful output should therefore make it easy to distinguish between:
- Confirmed information supported by the requirements or evidence
- Assumptions that still need validation
- Gaps and ambiguities requiring further investigation
- Questions that need to go back to stakeholders
- Draft user stories and acceptance criteria that can be progressively refined
Remember
GenAI can help you analyse, structure and challenge requirements, but it should not silently fill in missing information. As a Business Analyst, you still provide the context, validate assumptions, resolve ambiguities with stakeholders and decide when a requirement is sufficiently understood.
DSM Tip
The goal is not just write better user stories. It is to define clearer requirements with fewer hidden gaps before development begins.
