An agent team collaborates to derive optimal pricing through cost, competitive, value, and simulation analysis.
Execution Mode
Agent Team — 5 members communicate directly via SendMessage and perform cross-validation.
Agent Composition
| Agent | File | Role | Type |
|---|---|---|---|
| cost-analyst | .claude/agents/cost-analyst.md |
Cost structure, BEP, margin | general-purpose |
| competitive-analyst | .claude/agents/competitive-analyst.md |
Competitive pricing, positioning | general-purpose |
| value-assessor | .claude/agents/value-assessor.md |
Value-based pricing, WTP, segments | general-purpose |
| pricing-simulator | .claude/agents/pricing-simulator.md |
Scenarios, elasticity, P&L | general-purpose |
| pricing-reviewer | .claude/agents/pricing-reviewer.md |
Cross-validation, numerical consistency | general-purpose |
Workflow
Phase 1: Preparation (Performed directly by orchestrator)
- Extract from user input:
- Product/Service Information: Type, features, target customers
- Cost Data (optional): Known cost items
- Competitive Information (optional): Known competitors, prices
- Objectives: Revenue targets, market share targets, margin targets
- Create
_workspace/directory at the project root - Organize the input and save to
_workspace/00_input.md - If existing files are provided, copy them to
_workspace/and skip the corresponding Phase
Phase 2: Team Assembly and Execution
| Order | Task | Assigned To | Dependencies | Deliverable |
|---|---|---|---|---|
| 1a | Cost Analysis | cost-analyst | None | _workspace/01_cost_analysis.md |
| 1b | Competitive Pricing Analysis | competitive-analyst | None | _workspace/02_competitive_pricing.md |
| 2 | Value-Based Pricing | value-assessor | Tasks 1a, 1b | _workspace/03_value_pricing.md |
| 3 | Pricing Simulation | pricing-simulator | Tasks 1a, 1b, 2 | _workspace/04_pricing_simulation.md |
| 4 | Pricing Review | pricing-reviewer | Tasks 1a, 1b, 2, 3 | _workspace/05_review_report.md |
Tasks 1a (Cost) and 1b (Competitive) are executed in parallel. Both can start simultaneously as they have no initial dependencies.
Inter-team Communication Flow:
- cost-analyst completes → delivers price floor to competitive-analyst, delivers cost structure to value-assessor
- competitive-analyst completes → delivers competitive price range to value-assessor
- value-assessor completes → delivers WTP and pricing model recommendations to pricing-simulator
- pricing-simulator completes → delivers all documents to pricing-reviewer
- pricing-reviewer cross-validates all deliverables. If 🔴 Must Fix items are found, sends revision requests to the relevant agent → rework → re-validate (up to 2 times)
Phase 3: Integration and Final Deliverables
- Verify all files in
_workspace/ - Confirm that all 🔴 Must Fix items from the review report have been addressed
- Report the final summary to the user
Execution Modes by Request Scale
| User Request Pattern | Execution Mode | Agents Deployed |
|---|---|---|
| "Build a pricing strategy", "Full pricing analysis" | Full Pipeline | All 5 |
| "Just do a cost analysis" | Cost Mode | cost-analyst + reviewer |
| "Research competitive pricing" | Competitive Mode | competitive-analyst + reviewer |
| "Run a pricing simulation" (with data) | Simulation Mode | pricing-simulator + reviewer |
| "Review this pricing strategy" | Review Mode | reviewer only |
Data Transfer Protocol
| Strategy | Method | Purpose |
|---|---|---|
| File-based | _workspace/ directory |
Storing and sharing main deliverables |
| Message-based | SendMessage | Real-time key information transfer, revision requests |
Error Handling
| Error Type | Strategy |
|---|---|
| Insufficient cost data | Estimate based on industry benchmarks, tag as "estimates" |
| Competitor pricing undisclosed | Estimate using industry average range, tag as "undisclosed pricing" |
| Web search failure | Work from general knowledge, note "data limited" |
| Agent failure | Retry once → if still fails, proceed without that deliverable |
| 🔴 found in review | Send revision request to the relevant agent → rework → re-validate (up to 2 times) |
| Numerical contradiction found | Request numerical reconciliation from all related agents |
Test Scenarios
Normal Flow
Prompt: "I need to set the price for a B2B SaaS project management tool. Server costs are $15K/month, dev team labor is $230K/month. Competitors are Asana and Monday.com. Target ARR of $7.5M within 1 year." Expected Results:
- Cost: SaaS cost structure analysis, per-user unit cost, BEP
- Competitive: Asana/Monday pricing structure research, feature-price comparison
- Value: PM tool value drivers (time savings, collaboration efficiency), segment WTP
- Simulation: 3 scenario P&L, path to $7.5M ARR
- Review: All figures cross-validated
Existing File Utilization Flow
Prompt: "I've already done the cost analysis. Just do competitive analysis and pricing decision" + cost file attached Expected Results:
- Copy cost file to
_workspace/01_cost_analysis.md - Skip cost-analyst and deploy competitive-analyst + value-assessor + simulator + reviewer
Error Flow
Prompt: "Set the price for our app" Expected Results:
- Insufficient product info → ask follow-up questions about product type/target/features
- If only minimal info provided, present an industry-average-based price range
- Include note: "More precise analysis possible with additional data"
Agent Extension Skills
Extension skills that enhance agent domain expertise:
| Skill | File | Target Agents | Role |
|---|---|---|---|
| psm-analyzer | .claude/skills/psm-analyzer/skill.md |
value-assessor, pricing-simulator | Van Westendorp PSM 4 questions, intersection derivation, Gabor-Granger supplement, segment analysis |
| price-elasticity-calculator | .claude/skills/price-elasticity-calculator/skill.md |
pricing-simulator, competitive-analyst | PED/XED formulas, optimal price derivation, scenario simulation, price increase strategy |