agentic-data-stack-community/data-core/workflows/simple/greenfield/simple-analytics-project.yaml

AI Agentic Data Stack Framework - Community Edition. Open source data engineering framework with 4 core agents, essential templates, and 3-dimensional quality validation.

workflow:
  id: simple-analytics-project
  name: Simple Analytics Dashboard Project
  description: "Small-scale analytics project: Create a customer segmentation dashboard from existing CRM data"
  complexity: simple
  project_size: small
  duration: "2-4 weeks"
  team_size: "2-3 people"
  
metadata:
  use_case: "Business Intelligence Dashboard"
  industry: "E-commerce/Retail"
  data_volume: "< 1M records"
  data_sources: 1
  stakeholders: 3
  compliance_level: basic
  
context:
  business_scenario: |
    A small e-commerce company wants to understand their customer base better.
    They have customer data in their CRM system and want to create segments
    for targeted marketing campaigns. The marketing team needs a simple 
    dashboard showing customer segments, purchase patterns, and basic metrics.
    
  success_criteria:
    - Customer segmentation based on purchase behavior
    - Interactive dashboard with key metrics
    - Automated daily data refresh
    - Marketing team can self-serve insights
    
  constraints:
    - Single data source (CRM database)
    - Limited technical team (1 data analyst, 1 developer)
    - Small budget for tools and infrastructure
    - 4-week deadline for campaign launch

agents_involved:
  - data-product-manager
  - data-analyst
  - data-quality-engineer

workflow_stages:
  - stage: project_initiation
    name: "Project Kickoff and Planning"
    duration: "2 days"
    description: "Define requirements, scope, and initial planning"
    
    tasks:
      - task: stakeholder_alignment
        agent: data-product-manager
        description: "Meet with marketing team to understand requirements"
        deliverables:
          - Business requirements document
          - Success criteria definition
          - Timeline and resource planning
        activities:
          - "Interview marketing stakeholders"
          - "Define customer segmentation goals"
          - "Identify key metrics and KPIs"
          - "Establish project timeline and milestones"
        
      - task: data_discovery
        agent: data-analyst
        description: "Explore CRM data to understand structure and quality"
        deliverables:
          - Data profiling report
          - Initial data quality assessment
          - Segmentation feasibility analysis
        activities:
          - "Connect to CRM database"
          - "Profile customer and transaction data"
          - "Identify data quality issues"
          - "Assess segmentation variables availability"
        
      - task: technical_planning
        agent: data-analyst
        description: "Plan technical approach and tool selection"
        deliverables:
          - Technical architecture overview
          - Tool selection rationale
          - Development environment setup plan
        activities:
          - "Evaluate dashboard tools (Tableau, Power BI, etc.)"
          - "Plan data extraction and transformation approach"
          - "Design simple ETL process"
          - "Set up development environment"

  - stage: data_preparation
    name: "Data Analysis and Preparation"
    duration: "1 week"
    description: "Clean data and develop segmentation logic"
    
    tasks:
      - task: data_cleaning
        agent: data-analyst
        description: "Clean and prepare CRM data for analysis"
        deliverables:
          - Cleaned dataset
          - Data transformation scripts
          - Data quality report
        activities:
          - "Handle missing values and duplicates"
          - "Standardize data formats"
          - "Create calculated fields for analysis"
          - "Document data transformation rules"
        quality_gates:
          - "Data completeness > 95%"
          - "No duplicate customer records"
          - "All required fields populated"
          - "Data types properly formatted"
        
      - task: segmentation_development
        agent: data-analyst
        description: "Develop customer segmentation logic"
        deliverables:
          - Customer segments definition
          - Segmentation algorithm/rules
          - Segment validation report
        activities:
          - "Analyze purchase behavior patterns"
          - "Define segmentation criteria (RFM, demographics, etc.)"
          - "Create segmentation rules or model"
          - "Validate segments with business stakeholders"
        quality_gates:
          - "Segments are mutually exclusive"
          - "Each segment has meaningful business interpretation"
          - "Segment sizes are actionable for marketing"
          - "Stakeholder approval of segment definitions"

  - stage: dashboard_development
    name: "Dashboard Creation and Testing"
    duration: "1 week"
    description: "Build and test the analytics dashboard"
    
    tasks:
      - task: dashboard_design
        agent: data-analyst
        description: "Design and build interactive dashboard"
        deliverables:
          - Interactive customer segmentation dashboard
          - Dashboard documentation
          - User guide for marketing team
        activities:
          - "Create dashboard mockup and get approval"
          - "Build visualizations for each customer segment"
          - "Add filters and interactive elements"
          - "Implement key metrics and KPIs"
        quality_gates:
          - "Dashboard loads within 5 seconds"
          - "All visualizations display correctly"
          - "Interactive filters work as expected"
          - "Data refreshes without errors"
        
      - task: data_pipeline_setup
        agent: data-analyst
        description: "Set up automated data refresh pipeline"
        deliverables:
          - Automated ETL pipeline
          - Data refresh schedule
          - Pipeline monitoring setup
        activities:
          - "Create daily data extraction job"
          - "Set up data transformation pipeline"
          - "Configure dashboard data source refresh"
          - "Implement basic error handling and alerts"
        quality_gates:
          - "Pipeline runs successfully without manual intervention"
          - "Data refreshes complete within 1 hour"
          - "Error notifications sent to appropriate team members"
          - "Data lineage is documented"

  - stage: validation_deployment
    name: "Validation and Deployment"
    duration: "3-5 days"
    description: "User acceptance testing and production deployment"
    
    tasks:
      - task: user_acceptance_testing
        agent: data-product-manager
        description: "Conduct UAT with marketing team"
        deliverables:
          - UAT results report
          - Bug fixes and improvements
          - Final stakeholder approval
        activities:
          - "Train marketing team on dashboard usage"
          - "Conduct hands-on testing sessions"
          - "Gather feedback and implement minor changes"
          - "Get formal sign-off from stakeholders"
        quality_gates:
          - "Marketing team can navigate dashboard independently"
          - "All critical functionality works correctly"
          - "Performance meets user expectations"
          - "Stakeholders approve for production use"
        
      - task: production_deployment
        agent: data-analyst
        description: "Deploy to production and monitor"
        deliverables:
          - Production deployment
          - Monitoring setup
          - Handover documentation
        activities:
          - "Deploy dashboard to production environment"
          - "Set up production monitoring and alerts"
          - "Create operational documentation"
          - "Conduct knowledge transfer to support team"
        quality_gates:
          - "Production deployment successful"
          - "All security and access controls configured"
          - "Monitoring and alerting functional"
          - "Support documentation complete"

project_deliverables:
  primary:
    - "Interactive customer segmentation dashboard"
    - "Automated daily data pipeline"
    - "Customer segment definitions and business rules"
    
  supporting:
    - "Data profiling and quality report"
    - "Technical documentation"
    - "User training materials"
    - "Operational procedures"

technical_stack:
  data_source: "CRM Database (PostgreSQL/MySQL)"
  etl_tool: "SQL scripts + scheduled jobs"
  dashboard_tool: "Power BI/Tableau Public"
  scheduling: "Database scheduler or simple cron jobs"
  
quality_framework:
  data_quality:
    completeness: "> 95% for key fields"
    accuracy: "Customer data validated against business rules"
    consistency: "No duplicate customer records"
    timeliness: "Data refreshed daily before 8 AM"
    
  technical_quality:
    performance: "Dashboard loads within 5 seconds"
    availability: "99% uptime during business hours"
    usability: "Marketing team can use without technical support"
    
  business_quality:
    relevance: "Segments actionable for marketing campaigns"
    insights: "Clear differentiation between customer segments"
    adoption: "Marketing team actively uses dashboard weekly"

risk_management:
  technical_risks:
    - risk: "CRM database performance impact"
      mitigation: "Schedule data extraction during off-peak hours"
      probability: low
      impact: medium
      
    - risk: "Dashboard tool licensing costs"
      mitigation: "Use free/open-source alternatives if needed"
      probability: medium
      impact: low
      
  business_risks:
    - risk: "Marketing team adoption resistance"
      mitigation: "Involve team in design process and provide training"
      probability: low
      impact: high
      
    - risk: "Changing requirements during development"
      mitigation: "Implement core features first, enhancements later"
      probability: medium
      impact: medium

success_metrics:
  technical:
    - "Zero critical bugs in production after 1 week"
    - "Data pipeline 100% success rate for first month"
    - "Dashboard response time < 5 seconds"
    
  business:
    - "Marketing team uses dashboard at least 3x per week"
    - "Customer segmentation drives at least one campaign"
    - "Positive feedback from marketing stakeholders"
    
  adoption:
    - "100% of marketing team trained on dashboard"
    - "Self-service usage without IT support requests"
    - "Request for additional features/expansions"

lessons_learned_template:
  what_worked_well:
    - "Simple scope enabled quick delivery"
    - "Close collaboration with business stakeholders"
    - "Early data discovery prevented major surprises"
    
  challenges_faced:
    - "Data quality issues required additional cleaning time"
    - "Tool selection required balancing features vs. cost"
    - "Business stakeholder availability for feedback sessions"
    
  improvements_for_next_time:
    - "Allocate more time for data quality assessment"
    - "Create data dictionary upfront"
    - "Set up regular check-ins with stakeholders"
    
  recommendations:
    - "Consider this pattern for other simple analytics projects"
    - "Document reusable components for future projects"
    - "Plan for scaling if project proves successful"