---

name: ai-portfolio-optimizer description: Use when balancing AI investments across initiatives. Use during planning cycles. Produces portfolio analysis, resource allocation, and investment recommendations.

AI Portfolio Optimizer

Overview

Optimize AI investments across the portfolio of initiatives. Balance innovation, efficiency, and maintenance investments to maximize overall value.

Core principle: Organizations have limited AI resources. Strategic allocation across a balanced portfolio delivers more value than ad-hoc investment.

When to Use

• Annual/quarterly AI planning
• Resource allocation decisions
• Portfolio reviews
• Investment prioritization

Output Format

ai_portfolio:
  period: "[Planning period]"
  date: "[YYYY-MM-DD]"
  
  current_state:
    total_investment: "[$]"
    distribution:
      - category: "[Category]"
        allocation: "[%]"
        initiatives: "[N]"
    
    initiatives:
      - id: "[ID]"
        name: "[Name]"
        category: "[Innovation | Efficiency | Maintenance]"
        investment: "[$]"
        status: "[Active | Planned | Complete]"
        value_delivered: "[$]"
        health: "[Green | Amber | Red]"
  
  analysis:
    balance:
      current: "[Assessment of current balance]"
      recommendation: "[Target allocation]"
    
    value_realization:
      on_track: "[N initiatives]"
      at_risk: "[N initiatives]"
      underperforming: "[N initiatives]"
    
    resource_utilization:
      capacity: "[Available resources]"
      allocated: "[Committed]"
      available: "[Remaining]"
    
    gaps:
      - "[Strategic gap 1]"
  
  optimization:
    recommended_changes:
      - initiative: "[ID]"
        current: "[Current investment]"
        recommended: "[New investment]"
        rationale: "[Why]"
    
    new_investments:
      - opportunity: "[Description]"
        category: "[Category]"
        investment: "[$]"
        expected_value: "[$]"
    
    sunset_candidates:
      - initiative: "[ID]"
        reason: "[Why retire]"
        reallocation: "[Where to redirect]"
  
  target_portfolio:
    allocation:
      innovation: "[%]"
      efficiency: "[%]"
      maintenance: "[%]"
    
    initiatives:
      - id: "[ID]"
        investment: "[$]"
        priority: "[1-N]"
  
  recommendation:
    summary: "[Executive summary]"
    key_actions:
      - "[Action 1]"

Portfolio Categories

Innovation

innovation:
  definition: "New AI capabilities, exploratory, higher risk"
  examples:
    - "New AI products"
    - "Novel use cases"
    - "Emerging technology exploration"
  
  characteristics:
    - "Higher uncertainty"
    - "Longer time to value"
    - "Potential for high returns"
    - "Learning opportunities"
  
  target_allocation: "20-30%"

Efficiency

efficiency:
  definition: "AI to improve existing processes and products"
  examples:
    - "Automation of manual tasks"
    - "Cost reduction initiatives"
    - "Productivity improvements"
  
  characteristics:
    - "Moderate risk"
    - "Measurable ROI"
    - "Builds on existing capabilities"
  
  target_allocation: "40-50%"

Maintenance

maintenance:
  definition: "Keep existing AI systems running and current"
  examples:
    - "Model retraining"
    - "Infrastructure updates"
    - "Technical debt"
    - "Security patches"
  
  characteristics:
    - "Lower risk"
    - "Necessary investment"
    - "Avoids degradation"
  
  target_allocation: "20-30%"

Portfolio Balancing

Healthy vs Unhealthy Portfolio

healthy_signs:
  - "Balanced across categories"
  - "Pipeline of innovation feeding future"
  - "Efficiency initiatives delivering ROI"
  - "Maintenance preventing degradation"
  - "Resources aligned to priorities"

unhealthy_signs:
  - "All resources on maintenance (no innovation)"
  - "Only innovation (tech debt growing)"
  - "Scattered investments (no focus)"
  - "Over-committed resources"

Rebalancing Triggers

Trigger	Action
Innovation > 40%	More risk than sustainable
Maintenance > 40%	Starving future investment
Efficiency underperforming	Review or reallocate
Major strategy shift	Realign to new priorities

Prioritization Framework

Scoring Criteria

scoring:
  strategic_alignment:
    weight: 30%
    scale: "1-5"
  
  value_potential:
    weight: 25%
    scale: "ROI or $ value"
  
  feasibility:
    weight: 20%
    scale: "1-5 (data, tech, org readiness)"
  
  risk:
    weight: 15%
    scale: "1-5 (lower is less risky)"
  
  urgency:
    weight: 10%
    scale: "1-5 (competitive, regulatory drivers)"

Priority Matrix

              High Strategic Alignment
                        │
    ┌───────────────────┼───────────────────┐
    │  Do eventually    │  Must do          │
    │  (lower priority) │  (highest priority)│
Low ├───────────────────┼───────────────────┤High
Value│  Don't do         │  Quick wins       │Value
    │  (deprioritize)   │  (do if capacity) │
    └───────────────────┼───────────────────┘
                        │
              Low Strategic Alignment

Resource Optimization

Capacity Planning

capacity:
  roles:
    - role: "Data Scientist"
      available_fte: "[N]"
      allocated_fte: "[N]"
      utilization: "[%]"
    
    - role: "ML Engineer"
      available_fte: "[N]"
      allocated_fte: "[N]"
      utilization: "[%]"
  
  constraints:
    - "[Constraint 1]"
  
  recommendations:
    - "[Hire, reallocate, or defer based on gaps]"

Portfolio Review Cadence

Review	Frequency	Focus
Executive	Quarterly	Strategic alignment, major decisions
Portfolio	Monthly	Health, progress, risks
Initiative	Weekly	Execution, blockers

Checklist

• Current portfolio documented
• Initiatives categorized
• Value delivery assessed
• Balance analyzed
• Optimization opportunities identified
• Resource capacity considered
• Recommendations prioritized
• Stakeholder alignment