Pattern: Parallel Workers | Team size: 5 agents
This team rapidly explores multiple Officeworks business domains in parallel to identify the strongest demo use case for Fabric IQ and Copilot Studio. The parallel structure works because merchandising, supply chain, customer service, and store operations can be assessed independently, then compared for business impact and demo feasibility.
I have a demo with Office works to showcase how muti agent approarch using fabric iq and copilot stuiduo is something they need but i dont know what usecase i build and show that it become a compelling business demo rather than a techncal demo
Create an agent team to discover, compare, and recommend the most compelling business demo use case for Officeworks that showcases a multi-agent approach using Microsoft Fabric IQ and Copilot Studio as a business-value demo, not a technical demo. Project name: officeworks_multi_agent_demo_discovery Business context: The user has an upcoming demo with Officeworks. They want to show why a multi-agent approach using Fabric IQ and Copilot Studio is something Officeworks needs. The user does not yet know which use case to build. The goal is to identify a compelling, executive-relevant demo scenario that feels like an Officeworks business solution, not a generic technical showcase. Officeworks context to assume: - Officeworks is a large Australian retailer focused on office supplies, technology, furniture, print/copy services, education supplies, business supplies, and work-from-home products. - It serves consumers, students, small businesses, enterprise/business buyers, schools, teachers, and retail store customers. - Relevant domains may include merchandising, inventory, store operations, online customer experience, click-and-collect, print/copy services, business account support, product advice, returns, supplier issues, and seasonal demand such as back-to-school. - The demo should be suitable for business stakeholders such as retail executives, operations leaders, digital/customer leaders, merchandising leaders, and IT/data leaders. - Do not assume access to private Officeworks data. Use plausible synthetic/example data sources and clearly label assumptions. Technology context: - Copilot Studio can be positioned as the conversational and orchestration layer for role-specific agents and customer/staff-facing copilots. - Fabric IQ can be positioned as the intelligence/business-context layer using unified data, semantic models, metrics, business entities, insights, and contextual grounding across retail operations. - The demo should show multiple specialized agents collaborating, each with clear responsibilities, not one chatbot answering generic questions. - The story should emphasize business outcomes: increased sales, lower service cost, better stock availability, faster store response, improved customer satisfaction, reduced manual work, better decisions, and trusted data. Team pattern: Parallel Workers. Each agent works independently on a specific angle first, then the team compares findings and synthesizes the best demo use case. Agents: 1. Retail Pain Point Analyst 2. Customer Experience Strategist 3. Store Operations Specialist 4. Data and Fabric IQ Mapper 5. Demo Value Scorer Global instructions for all agents: - Work in parallel during Phase 1. - Be concrete. Avoid vague statements like “improve efficiency” unless tied to a workflow, stakeholder, measurable outcome, and demo moment. - Focus on use cases that can be demonstrated in 8–12 minutes. - Prefer use cases where multiple agents clearly coordinate and where Fabric IQ adds business grounding beyond a chatbot. - Include assumptions explicitly. - Use Officeworks-relevant examples, products, personas, and scenarios. - Distinguish between “business demo narrative” and “technical implementation detail.” - Do not write code. - Create all outputs under outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/ - Use markdown files only. Required workflow and dependencies: Phase 1 — Parallel discovery All five agents begin at the same time and independently create their own analysis files. This phase must complete before Phase 2 begins. Phase 2 — Cross-agent challenge and refinement After all Phase 1 files are complete, each agent must read the other agents’ outputs and add a challenge/refinement section to its own final file or create the required challenge file. This task must complete before Phase 3 begins. Phase 3 — Scoring and shortlist Demo Value Scorer consolidates candidate use cases into a scored shortlist. This task must complete before Phase 4 begins. Phase 4 — Final synthesis and recommendation Create a final recommendation pack that identifies the best use case to build, the demo story, the agent architecture, Fabric IQ role, Copilot Studio role, business value, and next steps. This must be the final step. Deliverable overview: Create exactly these files: 1. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/01_retail_pain_points.md 2. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/02_customer_experience_opportunities.md 3. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/03_store_operations_workflows.md 4. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/04_fabric_iq_data_mapping.md 5. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/05_candidate_usecase_scorecard.md 6. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/06_cross_agent_challenges.md 7. outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/07_final_recommendation_pack.md Agent 1: Retail Pain Point Analyst Objective: Identify high-value Officeworks business problems where multi-agent automation could reduce cost, improve service, increase sales, reduce operational friction, or improve executive decision-making. Create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/01_retail_pain_points.md Structure the file exactly as follows: # Retail Pain Point Analysis ## Executive Summary Write 120–180 words summarizing the strongest Officeworks business pain themes and why they matter for a demo. ## Assumptions List 5–8 explicit assumptions about Officeworks operations, data availability, business priorities, and demo constraints. ## Top Business Pain Areas Provide exactly 8 pain areas. For each use this format: ### Pain Area [number]: [name] - Business problem: - Affected stakeholders: - Likely frequency: - Business impact: - Why current process may be painful: - Why a multi-agent approach helps: - Fabric IQ relevance: - Copilot Studio relevance: - Demo potential: High/Medium/Low - Example Officeworks scenario: ## Best Candidate Use Cases from Pain Analysis List exactly 5 candidate use cases. For each include: - Use case name - Primary business outcome - Main persona - Supporting personas - What the demo would show in 60 seconds - Why this is more than a chatbot ## Risks and Watchouts List 6 risks that could make a use case less compelling or harder to demo. ## Recommended Pain-Led Shortlist Rank the top 3 use cases from this agent’s perspective and explain each in 80–120 words. Agent 2: Customer Experience Strategist Objective: Find customer-facing Officeworks moments where Copilot Studio agents and multi-agent collaboration can create a visible, emotionally compelling, and commercially valuable improvement. Create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/02_customer_experience_opportunities.md Structure the file exactly as follows: # Customer Experience Opportunity Analysis ## Executive Summary Write 120–180 words describing the most compelling customer-facing demo opportunities. ## Customer Personas Define exactly 6 personas: 1. Small business owner 2. Parent preparing for back-to-school 3. Student 4. Teacher/school administrator 5. Remote worker/home office buyer 6. Business account/procurement buyer For each persona include: - Goals - Frustrations - High-value moments - Demo-relevant data signals ## Customer Journey Pain Points Create a table with exactly 10 rows and these columns: | Journey Moment | Customer Pain | Officeworks Impact | Multi-Agent Opportunity | Demo Visibility | ## Customer-Facing Use Case Concepts Provide exactly 6 concepts. For each include: - Concept name - Customer story - Agents involved - Data needed - Fabric IQ contribution - Copilot Studio contribution - Business value - 90-second demo moment - Potential executive reaction ## Best Customer Demo Candidate Choose 1 best candidate and write: - Candidate name - Why customers care - Why Officeworks executives care - Why multi-agent is necessary - How the demo can create a “wow” moment - What synthetic data would be needed ## Customer Experience Watchouts List 5 watchouts, including privacy, accuracy, channel fit, escalation, and over-personalization. Agent 3: Store Operations Specialist Objective: Evaluate Officeworks in-store workflows such as stock checks, product advice, returns, print/copy service issues, click-and-collect, staff assistance, and manager escalations for demo suitability. Create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/03_store_operations_workflows.md Structure the file exactly as follows: # Store Operations Workflow Analysis ## Executive Summary Write 120–180 words summarizing the strongest in-store and staff-facing opportunities. ## Store Workflow Inventory List exactly 10 workflows. For each include: - Workflow name - Current likely process - Staff pain point - Customer pain point - Manager pain point - Data involved - Multi-agent opportunity - Demo suitability: High/Medium/Low ## Deep Dive: Top 4 Store Workflows For exactly 4 workflows, provide: ### Workflow [number]: [name] - Trigger: - Frontline staff action: - Agent collaboration sequence: - Fabric IQ grounding: - Copilot Studio interaction: - Business outcome: - Demo scene: - Failure/escalation path: ## Best Store Operations Demo Candidate Choose 1 best store operations use case and include: - Candidate name - Target user - Why it matters in stores - Why it shows multi-agent value - Why it is feasible to demo with synthetic data - How it links to executive metrics ## Store Operations Risks List exactly 6 risks and mitigations. Agent 4: Data and Fabric IQ Mapper Objective: Map likely Officeworks data sources to Fabric IQ capabilities and explain how Fabric IQ can make the demo feel grounded in business context, metrics, entities, and trusted insights rather than a standalone chatbot. Create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/04_fabric_iq_data_mapping.md Structure the file exactly as follows: # Fabric IQ and Data Mapping ## Executive Summary Write 120–180 words explaining how Fabric IQ should be positioned in the demo. ## Likely Officeworks Data Sources Create a table with exactly 12 rows and these columns: | Data Source | Example Fields | Owner Function | Refresh Need | Demo Use | Synthetic Data Feasibility | Include likely sources such as POS sales, product catalogue, inventory, online orders, click-and-collect, customer service, returns, supplier/lead time, promotions, store roster/tasks, print/copy jobs, and business account data. ## Business Entities and Metrics Define exactly 10 business entities and exactly 12 metrics. For each business entity include: - Entity name - Description - Relationships to other entities - Demo relevance For each metric include: - Metric name - Definition - Business owner - Demo relevance ## Fabric IQ Capabilities to Showcase List exactly 8 capabilities. For each include: - Capability name - Business explanation - Demo example - Why it matters to trust/adoption ## Data-to-Agent Grounding Map Create a table with exactly 8 rows: | Agent or Copilot | Questions It Answers | Data/Metric Grounding | Fabric IQ Role | Output to User | ## Recommended Demo Data Model Write 250–350 words describing a minimal synthetic data model for the recommended demo. If no recommendation is final yet, propose a flexible model that supports inventory, customer journey, store operations, and sales scenarios. ## Data Risks and Mitigations List exactly 6 risks and mitigations. Agent 5: Demo Value Scorer Objective: Score candidate use cases by business value, executive appeal, technical feasibility, demo clarity, multi-agent necessity, Fabric IQ relevance, Copilot Studio fit, and wow factor. Phase 1 task: Review likely use cases independently using general Officeworks context and create an initial scoring approach in: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/05_candidate_usecase_scorecard.md After Phase 1 files from Agents 1–4 are complete, update this same file with the consolidated shortlist and scoring. This update must complete before Phase 4 begins. Structure the file exactly as follows: # Candidate Use Case Scorecard ## Scoring Method Define the scoring method with exactly 8 criteria: 1. Business value 2. Executive appeal 3. Demo clarity 4. Multi-agent necessity 5. Fabric IQ relevance 6. Copilot Studio fit 7. Technical feasibility with synthetic data 8. Wow factor For each criterion include: - Definition - 1-point meaning - 3-point meaning - 5-point meaning ## Initial Candidate Pool List 10–14 candidate use cases with one-sentence descriptions. ## Consolidated Candidate Shortlist After reading Agents 1–4, select exactly 6 shortlisted use cases. For each include: - Candidate name - Source agents that supported it - Business problem - Demo user - Agents involved - Fabric IQ role - Copilot Studio role - Key demo moment ## Scorecard Table Create a table with exactly 6 rows and these columns: | Use Case | Business Value | Executive Appeal | Demo Clarity | Multi-Agent Necessity | Fabric IQ Relevance | Copilot Studio Fit | Feasibility | Wow Factor | Total /40 | Rank | Use 1–5 scores for each criterion. ## Scoring Rationale For each of the 6 shortlisted use cases, write 80–120 words explaining the score and tradeoffs. ## Recommended Winner and Runner-Up Choose: - Winner - Runner-up - Why the winner beats the runner-up - What would change the recommendation Cross-agent collaboration file After all Phase 1 files are complete and before final synthesis, create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/06_cross_agent_challenges.md This file must explicitly compare perspectives and challenge assumptions. Structure the file exactly as follows: # Cross-Agent Challenges and Refinements ## Areas of Agreement List exactly 6 points where agents agree. ## Areas of Tension List exactly 6 tensions. For each include: - Tension: - Agents likely to disagree: - Why it matters: - Resolution: ## Assumption Challenges List exactly 8 assumptions that must be tested before building the demo. For each include: - Assumption - Why it may be wrong - Demo-safe fallback ## Use Case Refinements from Collaboration List exactly 5 refinements that improve the final demo recommendation. For each include: - Original idea - Challenge raised - Refined version - Benefit of refinement ## Final Alignment Statement Write 150–220 words stating what the team now agrees is the best direction and why. Final synthesis and recommendation After all prior files are complete, create: outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/07_final_recommendation_pack.md This is the final deliverable. It must synthesize all agent outputs into a practical recommendation the user can use to decide what demo to build. Structure the file exactly as follows: # Officeworks Multi-Agent Demo Recommendation Pack ## 1. Recommended Demo Use Case Write 200–300 words naming the recommended use case and explaining why it is the strongest business demo. The recommended use case should be framed as a business scenario, not a technology feature. It must clearly show: - Officeworks-specific relevance - Executive-level business value - Customer or staff pain - Why multiple agents are needed - Why Fabric IQ and Copilot Studio are both essential ## 2. Demo Title and One-Line Pitch Provide: - Demo title: maximum 12 words - One-line pitch: maximum 30 words - Executive hook: maximum 50 words ## 3. Target Audience List exactly 5 target stakeholders. For each include: - Role - What they care about - What the demo should make them believe ## 4. Business Storyboard for an 8–12 Minute Demo Create exactly 7 scenes. For each scene include: - Scene title - Persona on screen - Business situation - Agent interaction - Fabric IQ moment - Copilot Studio moment - Business outcome shown - Suggested duration The 7 scenes must form a coherent demo narrative from problem trigger to resolution to executive insight. ## 5. Multi-Agent Architecture Define exactly 5 demo agents. For each include: - Agent name - Primary role - Inputs - Decisions/actions - Outputs - Handoff to/from other agents - Why this cannot be replaced by a single generic chatbot ## 6. Fabric IQ Role in the Demo Write 250–350 words explaining Fabric IQ’s role as the business intelligence/context layer. Include: - Business entities - Metrics - Semantic grounding - Insights/recommendations - Trust and governance angle - How it makes the demo more credible ## 7. Copilot Studio Role in the Demo Write 200–300 words explaining Copilot Studio’s role as the conversational, orchestration, and action layer. Include: - User-facing copilot experiences - Agent orchestration - Escalations - Actions/workflows - Why it is visible and compelling in the demo ## 8. Synthetic Demo Data Required Create a table with exactly 10 rows: | Dataset | Example Records Needed | Key Fields | Used In Scene | Notes | ## 9. Business Value Case Create exactly 6 value levers. For each include: - Value lever - Current pain - Improvement shown in demo - Example metric - Executive relevance ## 10. Why This Is a Business Demo, Not a Technical Demo Write 180–250 words explaining how to present the demo so Officeworks sees operational and commercial value first, with technology as the enabler. ## 11. Build Plan for the Demo Provide a practical build plan with exactly 5 stages. For each stage include: - Stage name - Objective - Key tasks - Output - Dependency Make dependencies explicit, including “This task must complete before the next stage begins.” ## 12. Demo Risks and Mitigations List exactly 8 risks. For each include: - Risk - Impact - Mitigation ## 13. Final Recommendation Write 150–220 words with a decisive recommendation on what to build first, what to avoid, and how to position it in the Officeworks demo. Collaboration mechanics: - Agents must not simply duplicate each other. Each agent owns its domain and must produce distinct insights. - Agents must share candidate use case names and reconcile duplicates during Phase 2. - Agents must challenge at least one assumption from another domain in outputs/agent_teams_demo/officeworks_multi_agent_demo_discovery/06_cross_agent_challenges.md - Demo Value Scorer must use evidence from Agents 1–4, not only independent judgment. - Final synthesis must explicitly reference the scoring winner, the customer/store pain, and the Fabric IQ data grounding. - If agents disagree, resolve by prioritizing: business value first, demo clarity second, multi-agent necessity third, technical feasibility fourth, wow factor fifth. Quality bar: - The final recommendation must be specific enough that a demo builder can start creating synthetic data, agent roles, and a storyboard immediately. - Avoid generic retail language unless connected to Officeworks-specific scenarios such as back-to-school, click-and-collect, printer ink/technology advice, furniture availability, print/copy jobs, business account purchasing, returns, or store stock issues. - Make the recommended demo emotionally and commercially compelling. - Emphasize business outcomes over architecture diagrams. - Ensure every file uses the exact headings requested. - Ensure all files are internally consistent and point to the same final recommended use case by the end. - End the work with the final synthesis/review step in 07_final_recommendation_pack.md.
Create your own AI agent team at Build Agents Store. Describe your business problem and get specialized agent teams with ready-to-use prompts for Claude Code.