## Product Strategy Prompt: Apply GenAI to Enterprise Business Messaging Explain what generative AI is and how you would apply it to an enterprise business-messaging product. Choose one target industry segment, identify the most valuable use cases for that segment, and describe how you would evaluate success. ### Constraints & Assumptions - Assume the product supports business-to-customer conversations across channels such as chat, messaging apps, email, SMS, or in-app support. - Choose one industry segment and keep the use cases grounded in that segment's workflows. - The answer should cover user value, business value, technical feasibility, safety, privacy, and measurement. - Do not assume unlimited model accuracy; include human review, grounding, escalation, and compliance where needed. ### Clarifying Questions to Ask - Which customer segment are we targeting: SMBs, mid-market, or enterprise? - Which industry should we focus on first, and what workflows create the most messaging volume? - Are we optimizing for revenue conversion, support cost, response speed, customer satisfaction, or agent productivity? - What data sources are available for grounding: catalog, CRM, order status, knowledge base, policies, or conversation history? - What compliance requirements apply to the selected industry and channels? ### What a Strong Answer Covers - A plain definition of generative AI and why it matters in business messaging. - A specific target segment, such as retail/e-commerce, travel, financial services, healthcare, or telecom. - A prioritized use-case portfolio with clear user pain points and value. - A phased MVP that reduces risk before automating high-stakes conversations. - Success metrics across customer experience, agent productivity, business outcomes, quality, safety, and cost. - Risks such as hallucination, privacy leakage, brand voice drift, compliance errors, bias, and poor escalation. - A launch and experimentation plan with guardrails. ### Follow-up Questions - Which use case would you launch first and why? - How would you prevent hallucinated or non-compliant responses? - How would you price or package this feature for enterprise customers? - How would your plan change for a regulated industry? - What would you do if automation improves cost but lowers customer trust?