Value Stream Mapping
Create Lean value stream maps to visualize flow, identify waste, and design improvement opportunities. Based on Lean manufacturing principles adapted for knowledge work.
What is a Value Stream?
A value stream is the sequence of activities required to deliver value to a customer, from request to delivery. Value stream mapping (VSM) visualizes this flow to identify:
-
Value-adding activities - Steps that directly create customer value
-
Waste (Muda) - Steps that consume resources without adding value
-
Flow problems - Bottlenecks, delays, and inefficiencies
The 8 Wastes (TIMWOODS)
Waste Description Examples in Knowledge Work
Transportation Moving work between locations Handoffs between teams, tool switching
Inventory Unfinished work waiting Backlog, WIP, queued requests
Motion Unnecessary movement Context switching, searching for info
Waiting Idle time Approvals, blocked work, dependencies
Overproduction Making more than needed Unused features, premature optimization
Overprocessing Doing more than required Gold plating, excessive documentation
Defects Errors requiring rework Bugs, miscommunication, rework
Skills Underutilized talent Manual work that could be automated
Workflow
Phase 1: Define Scope
Step 1: Identify the Value Stream
Question Answer
What value are we delivering? [Customer outcome]
Who is the customer? [Internal/External customer]
Where does the stream start? [Trigger/Request]
Where does it end? [Value delivered]
Step 2: Set Boundaries
Value Stream Definition
Name: Customer Order Fulfillment Trigger: Customer places order End State: Customer receives product Scope: Order entry → Shipping (excludes manufacturing) Customer: External retail customer
Phase 2: Current State Map
Step 1: Walk the Process
Observe actual work (gemba walk):
-
Follow a real work item through the process
-
Document what actually happens, not what should happen
-
Note handoffs, delays, and workarounds
Step 2: Document Process Steps
For each step, capture:
Metric Definition How to Measure
Process Time (PT) Active work time Time actually working
Lead Time (LT) Total elapsed time Clock time start to finish
Wait Time (WT) Time waiting LT - PT
% Complete & Accurate (%C&A) First-time quality % not requiring rework
Step 3: Create Current State Map
Current State Value Stream
| Step | Owner | PT | LT | %C&A | Inventory | Notes |
|---|---|---|---|---|---|---|
| Order Entry | Sales | 15m | 2h | 85% | 50 orders | Manual entry |
| Credit Check | Finance | 10m | 8h | 95% | 30 orders | Batch processing |
| Inventory Alloc | Warehouse | 5m | 4h | 90% | 20 orders | System lookup |
| Pick & Pack | Warehouse | 30m | 6h | 92% | 40 orders | Manual process |
| Ship | Logistics | 10m | 24h | 98% | 100 orders | Carrier pickup |
Total Process Time: 70 minutes Total Lead Time: 44 hours Flow Efficiency: 2.7% (PT / LT)
Phase 3: Analyze Waste
Step 1: Calculate Flow Efficiency
Flow Efficiency = Process Time / Lead Time × 100%
Example: 70 min / 2640 min = 2.7%
Interpretation:
-
< 5%: Significant waste (typical for unoptimized processes)
-
5-15%: Moderate efficiency
-
15-25%: Good efficiency
25%: Excellent (rare for knowledge work)
Step 2: Identify Waste by Type
Waste Analysis
Waiting (45% of lead time)
- Credit check batch processing: 8 hours → could be real-time
- Carrier pickup schedule: 24 hours → could be on-demand
Inventory (150 orders WIP)
- Orders queue at each step
- No pull system in place
Defects (15% rework at order entry)
- Manual data entry errors
- Missing required fields
Overprocessing
- Full credit check for returning customers
- Duplicate data entry across systems
Step 3: Identify Bottlenecks
The constraint (bottleneck) limits the entire system:
Bottleneck Analysis
Primary Bottleneck: Credit Check
- Highest queue (30 orders)
- Batch processing creates 8-hour wait
- Blocks downstream flow
Secondary Bottleneck: Pick & Pack
- Manual process
- Variable cycle time
Phase 4: Future State Design
Step 1: Apply Lean Principles
Principle Current Problem Future State Solution
Flow Batch processing Continuous flow, single-piece
Pull Push based on forecast Pull based on demand
Perfection Accept defects, fix later Build quality in
Step 2: Design Improvements
Future State Design
Eliminate Waiting
- Real-time credit API vs. batch
- On-demand carrier pickup
Reduce Inventory
- WIP limits at each step
- Pull signals between steps
Improve Quality
- Validation at order entry
- Auto-populate from CRM
Automate Motion Waste
- Single system vs. multiple tools
- API integrations
Step 3: Create Future State Map
Future State Value Stream
| Step | Owner | PT | LT | %C&A | Inventory | Changes |
|---|---|---|---|---|---|---|
| Order Entry | Sales | 10m | 30m | 98% | 10 orders | Validation, auto-fill |
| Credit Check | System | 1m | 5m | 99% | 0 | Real-time API |
| Inventory Alloc | System | 1m | 5m | 99% | 0 | Automated |
| Pick & Pack | Warehouse | 25m | 2h | 96% | 15 orders | Better tooling |
| Ship | Logistics | 10m | 4h | 99% | 20 orders | On-demand pickup |
Target Process Time: 47 minutes (33% reduction) Target Lead Time: 6.7 hours (85% reduction) Target Flow Efficiency: 11.7% (4x improvement)
Phase 5: Implementation Roadmap
Step 1: Prioritize Improvements
Improvement Impact Effort Priority
Real-time credit API High (8h → 5m) Medium 1
Order entry validation Medium (15% → 2% errors) Low 2
On-demand carrier High (24h → 4h) High 3
Automated inventory Medium Medium 4
Step 2: Define Kaizen Events
Focused improvement workshops:
Kaizen Event: Credit Check Automation
Scope: Eliminate credit check batch processing Target: 8 hours → 5 minutes lead time Team: Finance, IT, Process Owner Duration: 1 week intensive Deliverables:
- API integration specification
- Process redesign
- Training materials
Output Formats
Narrative Summary
Value Stream Analysis Summary
Value Stream: [Name] Date: [ISO date] Analyst: value-stream-analyst
Current State Metrics
- Process Time: X minutes
- Lead Time: Y hours
- Flow Efficiency: Z%
- Primary Bottleneck: [Step]
Top Waste Categories
- Waiting (X% of LT): [Description]
- Inventory (X orders WIP): [Description]
- [Other waste]: [Description]
Future State Targets
- Lead Time Reduction: X%
- Flow Efficiency Target: Y%
- Quality Target: Z% first-time right
Recommended Actions
- [High priority improvement]
- [Medium priority improvement]
- [Lower priority improvement]
Structured Data (YAML)
value_stream: name: "Customer Order Fulfillment" version: "1.0" date: "{ISO-8601-date}" analyst: "value-stream-analyst"
boundaries: trigger: "Customer places order" end_state: "Customer receives product" customer: "External retail customer"
current_state: total_process_time_minutes: 70 total_lead_time_hours: 44 flow_efficiency_percent: 2.7 wip_total: 240
steps:
- name: "Order Entry"
owner: "Sales"
process_time_minutes: 15
lead_time_hours: 2
complete_accurate_percent: 85
inventory: 50
waste_types:
- type: defects
description: "Manual entry errors"
- type: motion
description: "Multiple system entry"
- name: "Credit Check"
owner: "Finance"
process_time_minutes: 10
lead_time_hours: 8
complete_accurate_percent: 95
inventory: 30
is_bottleneck: true
waste_types:
- type: waiting
description: "Batch processing"
waste_analysis: - type: waiting percent_of_lead_time: 45 root_causes: - "Batch processing" - "Scheduled pickups"
- type: inventory
total_wip: 240
root_causes:
- "No WIP limits"
- "Push system"
future_state: target_process_time_minutes: 47 target_lead_time_hours: 6.7 target_flow_efficiency_percent: 11.7
improvements:
- name: "Real-time credit API"
waste_eliminated: waiting
impact: high
effort: medium
lead_time_reduction_hours: 7.9
roadmap: - phase: 1 improvements: ["Real-time credit API", "Order validation"] duration_weeks: 4
- phase: 2
improvements: ["On-demand carrier", "Inventory automation"]
duration_weeks: 8
Mermaid Diagrams
Current State Flow:
flowchart LR subgraph Current State A[Order Entry<br/>PT: 15m | LT: 2h<br/>WIP: 50] --> B[Credit Check<br/>PT: 10m | LT: 8h<br/>WIP: 30] B --> C[Inventory<br/>PT: 5m | LT: 4h<br/>WIP: 20] C --> D[Pick & Pack<br/>PT: 30m | LT: 6h<br/>WIP: 40] D --> E[Ship<br/>PT: 10m | LT: 24h<br/>WIP: 100] end
style B fill:#ff9999
Future State Flow:
flowchart LR subgraph Future State A[Order Entry<br/>PT: 10m | LT: 30m<br/>WIP: 10] --> B[Credit<br/>API: 1m<br/>WIP: 0] B --> C[Inventory<br/>Auto: 1m<br/>WIP: 0] C --> D[Pick & Pack<br/>PT: 25m | LT: 2h<br/>WIP: 15] D --> E[Ship<br/>PT: 10m | LT: 4h<br/>WIP: 20] end
style B fill:#99ff99
style C fill:#99ff99
When to Use
Scenario Use Value Stream Mapping?
Process improvement Yes - identify waste
New system design Yes - design for flow
Cost reduction Yes - find inefficiencies
Lead time complaints Yes - find bottlenecks
Quality problems Partial - with root cause analysis
Greenfield project Maybe - limited current state
Integration
Upstream (Discovery)
-
process-modeling - Understand current process
-
capability-mapping - Link to business capabilities
-
stakeholder-analysis - Identify process owners
Downstream
-
Requirements - Improvement requirements
-
Systems design - Automation opportunities
-
Project planning - Kaizen event planning
Related Skills
-
journey-mapping
-
Customer experience perspective
-
capability-mapping
-
Capability view of value delivery
-
root-cause-analysis
-
Investigate bottleneck causes
-
prioritization
-
Prioritize improvement initiatives
-
process-modeling
-
Detailed BPMN process diagrams
-
estimation
-
Estimate improvement effort
-
benchmarking
-
Current vs target comparison
User-Facing Interface
When invoked directly by the user, this skill operates as follows.
Arguments
-
<process-description> : Description of the value stream or process to analyze
-
--mode : Analysis mode (default: guided )
-
full : Comprehensive analysis with current/future state maps (~12K tokens)
-
quick : Rapid waste identification (~4K tokens)
-
guided : Interactive value stream mapping (variable)
-
--output : Output format (default: both )
-
current : Current state map only
-
future : Future state map only
-
both : Both current and future state
-
--dir : Output directory (default: docs/analysis/ )
Execution Workflow
-
Parse Arguments - Extract process description, mode, and output format. If no description provided, ask the user what value stream to analyze (what value is delivered, what triggers the process, what marks completion).
-
Execute Based on Mode:
-
Full: Spawn the value-stream-analyst agent for end-to-end analysis with Process Time, Lead Time, %C&A metrics, TIMWOODS waste analysis, bottleneck identification (Theory of Constraints), and future state design with Kaizen event proposals.
-
Quick: Rapid identification of major steps, rough lead times, obvious waste (waiting, handoffs, rework), approximate flow efficiency, and 3-5 quick wins.
-
Guided: Interactive mapping walking through scope definition, step-by-step process documentation (PT, LT, %C&A, owner), metric calculation, waste identification by TIMWOODS category, bottleneck identification, and future state design.
-
Generate Output - Produce current state map, flow diagram (Mermaid flowchart), waste analysis, future state map with improvement targets, and prioritized improvement roadmap with Kaizen event proposals.
-
Save Results - Save to docs/analysis/value-stream-[name].md (or custom --dir ).
-
Suggest Follow-Ups - Recommend stakeholder-analysis for change management, journey-mapping for customer experience impact, and capability-mapping for process-capability alignment.
Version History
- v1.0.0 (2025-12-26): Initial release