The 2026 AI Playbook
for BA · FA · PO

Role: Senior Business Analyst preparing for a stakeholder discovery session.

Context: [Project name] — a [domain] system being built for [user group].
The primary stakeholder is [role, e.g. Head of Operations].

Task: Generate a structured interview guide with:
  • 5 opening questions to surface the business problem (not the solution)
  • 4 questions to uncover tacit knowledge and undocumented workarounds
  • 3 questions to identify constraints (regulatory, technical, organisational)
  • 2 closing questions to confirm understanding and agree next steps

Constraints: No leading questions. Avoid solution-first framing.
Each question should be open-ended and invite narrative.

Output format: Numbered list grouped by section, with a one-line "why this question" note per item.

Role: Functional Analyst writing machine-parseable requirements.

Context: Feature: [feature description in plain language]
System: [system name and brief description]
Actors: [list the actors involved]

Task: Convert the above into EARS-formatted requirements using the correct
template for each type:
  • Ubiquitous:  The [system] shall [action]
  • Event-driven: When [trigger], the [system] shall [response]
  • State-driven: While [state], the [system] shall [action]
  • Optional:     Where [feature] is included, the [system] shall [action]

Constraints: Each statement covers exactly one behaviour.
No compound "and/or" in a single requirement.
Quantify non-functionals (e.g. "within 500 ms", "≥ 99.5% uptime").

Output format: Table with columns: ID | EARS Type | Requirement | Rationale

Role: Product Owner prioritising the next sprint's candidates.

Context: Product: [product name]. Current sprint goal: [goal].
Stories for scoring: [paste story titles and one-line descriptions]

Task:
  1. Score each story on WSJF components (1–10 scale):
     • User/Business Value, Time Criticality, Risk Reduction, Job Size
     WSJF = (Value + Time + Risk) / Size
  2. Classify each story by Kano category:
     • Must-Have / Performance / Excitement / Indifferent
  3. Flag any story where Job Size > 5 as a split candidate.

Important: Use WSJF scores for relative ranking ONLY.
Job size estimates from AI run 10–20× too high for absolute planning.

Output format: Ranked table: Story | Value | Time | Risk | Size | WSJF | Kano | Split?

Role: BA / PO producing a Sprint Review summary.

Context:
Sprint: [sprint number/name]. Goal: [sprint goal]
Completed: [list stories done]
Incomplete: [list stories carried over, with reason]
Impediments: [list blockers encountered]
Metrics: Velocity [N] pts. Burn-up: [status]

Task: Produce THREE versions of the status report:
  A. Executive (≤ 5 bullet points) — business impact language, no jargon
  B. Steering / Sponsor (1 page) — decisions needed, risks, timeline impact
  C. Team retrospective input — honest, blameless, focused on process

Output format: Three labelled sections. Version A must lead with the sprint goal
outcome (achieved / partial / missed) in the first sentence.