Skip to main content
Sarah_Masters
Adobe Champion
Adobe Champion
Nov 13, 2026 1:21 PM UTC

From Business Request to PRD: An AI Workflow for Adobe AEP & AJO

[OVERVIEW] As a product manager working with Adobe Experience Platform and Adobe Journey Optimizer, I regularly translate marketing requests into product requirements that engineering teams can evaluate and build against. AI has become a useful part of that process, particularly for identifying missing requirements, researching Adobe capabilities, and turning business conversations into structured documentation. I've developed a three-stage workflow using ChatGPT to help with requirements discovery, PRD generation, and quality review. The goal isn't to have AI make product decisions. It's to reduce the manual effort of documenting requirements while improving their clarity, consistency, and testability. [TRYABLE] Stage 1: Requirements Discovery Start with a business request, even if it's incomplete. ChatGPT reviews the request, identifies gaps, researches relevant Adobe capabilities using official documentation when available, and asks up to five targeted clarification questions. This helps surface important considerations such as data availability, audience qualification, identity, consent, activation, personalization, timing, and measurement before the PRD is written. Stage 2: PRD Generation Once the discovery questions are answered, ChatGPT creates a structured PRD containing: - Business context and objectives - Use cases and relevant delivery boundaries - Numbered requirements with testable acceptance criteria - Business success measures - Dependencies, risks, and open questions - Engineering discovery considerations The focus is on defining what the business needs and why, without prescribing technical implementation decisions. Stage 3: Quality Review Before finalizing the document, ChatGPT reviews its own output for ambiguous requirements, missing acceptance criteria, duplication, unsupported assumptions, and unnecessary technical prescriptions. It corrects issues where the intended meaning is clear and flags unresolved decisions rather than inventing answers. The result is a reviewed PRD that can serve as a starting point for engineering discovery and stakeholder alignment. Reusable Prompt: Act as a Senior MarTech Product Manager with expertise in Adobe Experience Platform (AEP), Real-Time CDP, and Adobe Journey Optimizer (AJO). Help me turn a business request into a clear, concise Product Requirements Document (PRD) suitable for engineering discovery. Your responsibility is to define what the business needs and why, not prescribe how engineering should implement it. Follow this three-stage workflow. Stage 1: Requirements Discovery 1. Summarize the business problem and intended outcome. 2. Identify confirmed requirements, constraints, and dependencies. 3. Identify material gaps involving data, identity, consent, audience qualification, activation, personalization, timing, measurement, and operational needs, where relevant. 4. Research applicable Adobe functionality using official documentation when available. 5. Ask up to five high-impact clarification questions before drafting. Distinguish confirmed requirements, verified Adobe capabilities, unverified assumptions, and open decisions. Do not proceed until I answer the questions or ask you to continue with unresolved items. Stage 2: PRD Generation Create a concise PRD using the following structure: 1. Overview & Business Context - Business problem, objectives, and desired outcome. 2. Use Cases & Delivery Boundaries - Relevant marketing use cases. - Include delivery boundaries only when necessary to clarify meaningful limitations or exclusions. 3. Requirements & Acceptance Criteria - Consolidated, numbered business and functional requirements. - Clear, observable, testable acceptance criteria. - Relevant dependencies or unresolved decisions. 4. Success Measures - Measurable business outcomes where known. - Distinguish success measures from functional acceptance criteria. 5. Dependencies, Risks & Open Questions - Material constraints, risks, unresolved decisions, and suggested owners. 6. Engineering Discovery - Technical considerations requiring investigation. - Applicable Adobe capabilities and limitations. - Questions for engineering and solution architects. Do not prescribe architecture, schemas, ingestion patterns, or integrations unless explicitly required. Stage 3: PRD Quality Review 1. Review the draft against these criteria: 2. Clarity: Are requirements unambiguous? 3. Testability: Can acceptance criteria be objectively verified? 4. Completeness: Are material business requirements missing? 5. Duplication: Are concepts unnecessarily repeated? 6. Accuracy: Are Adobe capabilities verified and correctly represented? 7. Scope control: Has anything been introduced that wasn't requested or approved? 8. Ownership: Are product requirements separated from implementation decisions? 9. Conciseness: Is every section adding value? Correct editorial issues, unnecessary repetition, and ambiguous language where the intended meaning is established. Do not invent missing requirements, business rules, SLAs, integrations, or success metrics. Flag decisions requiring stakeholder input rather than resolving them independently. Final Output Provide: 1. The complete reviewed PRD. 2. A brief quality review summary. 3. A handoff assessment identifying what is ready for engineering discovery and what remains unresolved. Keep the writing professional, direct, and concise. Scale the document to the complexity of the initiative. Prioritize requirements and acceptance criteria over lengthy background. Begin by asking me to describe the business request. [SETUP] Access to an AI assistant (ChatGPT, Claude, Gemini, etc.), a marketing business request, and Adobe Experience League documentation for capability validation. No Adobe environment access required. [IMPACT] Turn marketing requests into structured, engineering-ready PRDs with testable acceptance criteria, reducing manual documentation effort and catching requirement gaps before handoff. [LESSONS] Start with discovery, not documentation. Asking targeted questions before generating a PRD helps uncover requirements that might otherwise be missed. Don't assume the first draft is ready. AI can produce a polished document that still contains ambiguous or untestable requirements. A dedicated quality review makes a meaningful difference. Keep product and engineering responsibilities separate. AI should help define what the business needs, not make architecture or implementation decisions on behalf of engineering. Don't let AI fill in the blanks. Unverified Adobe capabilities, missing business rules, and unresolved decisions should be clearly flagged rather than presented as facts. Keep it concise. More documentation doesn't necessarily mean better requirements. Focus on what's needed for engineering to understand and evaluate the request.