Skip to content

Triple Diamond Workflow

Complete product development cycle from discovery to iteration

The Triple Diamond framework extends the traditional Double Diamond design model to encompass the full product lifecycle, adding dedicated phases for measurement and continuous improvement.

Workflow Metadata

Field Value
Workflow Triple Diamond
Command No dedicated command yet -- reference file directly
Skills All 25 phase skills across 6 phases
Phases Covered All 6 (Discover, Define, Develop, Deliver, Measure, Iterate)
Estimated Duration Weeks to months (comprehensive)
Prerequisite Inputs A new product or major initiative
Final Output Complete product development cycle from research to iteration

Overview

The Triple Diamond consists of six phases organized into three diamonds:

Diamond 1: Discovery & Definition
┌─────────────────────────────────────────┐
│                                         │
│    DISCOVER          DEFINE             │
│    (Diverge)         (Converge)         │
│                                         │
└─────────────────────────────────────────┘

Diamond 2: Development & Delivery
┌─────────────────────────────────────────┐
│                                         │
│    DEVELOP           DELIVER            │
│    (Diverge)         (Converge)         │
│                                         │
└─────────────────────────────────────────┘

Diamond 3: Measurement & Iteration
┌─────────────────────────────────────────┐
│                                         │
│    MEASURE           ITERATE            │
│    (Diverge)         (Converge)         │
│                                         │
└─────────────────────────────────────────┘
graph LR
    subgraph "Diamond 1: Discovery"
        A["Discover"] --> B["Define"]
    end
    subgraph "Diamond 2: Development"
        C["Develop"] --> D["Deliver"]
    end
    subgraph "Diamond 3: Measurement"
        E["Measure"] --> F["Iterate"]
    end
    B --> C
    D --> E
    F -.->|"next cycle"| A

When to Use

Use the Triple Diamond when:

  • Starting a new product or major initiative — need comprehensive discovery and validation
  • Building features with significant uncertainty — multiple unknowns about users or solutions
  • Establishing a new team or practice — want structured approach to build discipline
  • Working on high-stakes projects — cost of failure is high, need thorough validation

Consider a lighter approach (see feature-kickoff.md) when:

  • Requirements are well understood
  • Building incremental improvements
  • Time-constrained rapid development

Phase 1: Discover

Goal: Understand the problem space through research

Activities: - User research and interviews - Market and competitive analysis - Stakeholder mapping - Opportunity identification

Skills

Skill Description
discover-interview-synthesis Synthesize user research into actionable insights
discover-competitive-analysis Map competitive landscape and identify opportunities
discover-stakeholder-summary Document stakeholder needs and constraints

Key Outputs

  • Research synthesis with themes and patterns
  • Competitive landscape map
  • Stakeholder map with communication plan
  • List of opportunities to explore

Transition Criteria

Move to Define when: - [ ] Conducted research with target users (5+ interviews recommended) - [ ] Understand competitive landscape - [ ] Identified key stakeholders and their needs - [ ] Have clear opportunities to evaluate


Phase 2: Define

Goal: Frame the problem and form hypotheses

Activities: - Problem definition and scoping - Opportunity prioritization - Hypothesis formation - Jobs-to-be-done analysis

Skills

Skill Description
define-problem-statement Create clear problem framing with success criteria
define-hypothesis Define testable assumptions with metrics
define-opportunity-tree Map outcome-driven opportunities
define-jtbd-canvas Apply Jobs to be Done framework

Key Outputs

  • Problem statement with success criteria
  • Prioritized opportunity tree
  • Testable hypotheses
  • Clear understanding of the job to be done

Transition Criteria

Move to Develop when: - [ ] Problem is clearly defined and scoped - [ ] Have measurable success criteria - [ ] Hypotheses are specific and testable - [ ] Team is aligned on what problem to solve


Phase 3: Develop

Goal: Explore solution approaches

Activities: - Solution ideation and exploration - Technical spikes and feasibility analysis - Design exploration - Architecture decisions

Skills

Skill Description
develop-solution-brief Document proposed solution approach
develop-spike-summary Capture time-boxed exploration results
develop-adr Record architecture decisions
develop-design-rationale Document design decision reasoning

Key Outputs

  • Solution brief with trade-offs
  • Technical spike results
  • Architecture Decision Records
  • Design rationale documentation

Transition Criteria

Move to Deliver when: - [ ] Solution approach is validated - [ ] Key technical decisions are made and documented - [ ] Team is confident in feasibility - [ ] Major risks are identified and mitigated


Phase 4: Deliver

Goal: Specify, build, and ship

Activities: - Requirements specification - User story creation - Edge case analysis - Launch preparation

Skills

Skill Description
deliver-prd Write comprehensive product requirements
deliver-user-stories Generate user stories with acceptance criteria
deliver-edge-cases Document error states and recovery paths
deliver-launch-checklist Pre-launch validation checklist
deliver-release-notes User-facing release documentation

Key Outputs

  • Complete PRD
  • User stories with acceptance criteria
  • Edge case documentation
  • Launch checklist (signed off)
  • Release notes

Transition Criteria

Move to Measure when: - [ ] Feature is shipped to users - [ ] Instrumentation is in place - [ ] All launch checklist items complete - [ ] Release notes published


Phase 5: Measure

Goal: Validate with data

Activities: - Experiment design and execution - Analytics instrumentation - Dashboard creation - Results analysis

Skills

Skill Description
measure-experiment-design Design A/B tests and experiments
measure-instrumentation-spec Define event tracking requirements
measure-dashboard-requirements Specify analytics dashboard needs
measure-experiment-results Document experiment outcomes

Key Outputs

  • Experiment design with success criteria
  • Instrumentation specification
  • Dashboard with key metrics
  • Experiment results documentation

Transition Criteria

Move to Iterate when: - [ ] Experiments have reached statistical significance - [ ] Results are documented and communicated - [ ] Have clear learnings (positive or negative) - [ ] Data supports next steps decision


Phase 6: Iterate

Goal: Learn and improve continuously

Activities: - Team retrospectives - Lessons documentation - Backlog refinement - Pivot/persevere decisions

Skills

Skill Description
iterate-retrospective Facilitate team retrospectives
iterate-lessons-log Build organizational memory
iterate-refinement-notes Document backlog refinement outcomes
iterate-pivot-decision Framework for pivot/persevere decisions

Key Outputs

  • Retrospective action items
  • Lessons log entries
  • Refined backlog
  • Clear decision on next iteration

Cycle Continuation

Based on learnings, return to: - Discover — if fundamental assumptions were wrong - Define — if problem needs reframing - Develop — if solution needs significant changes - Deliver — if incremental improvements are needed


Full Skill Inventory

By Phase

Phase Skills (4)
Discover interview-synthesis, competitive-analysis, stakeholder-summary
Define problem-statement, hypothesis, opportunity-tree, jtbd-canvas
Develop solution-brief, spike-summary, adr, design-rationale
Deliver prd, user-stories, edge-cases, launch-checklist, release-notes
Measure experiment-design, instrumentation-spec, dashboard-requirements, experiment-results
Iterate retrospective, lessons-log, refinement-notes, pivot-decision

By Category

Category Skills
research interview-synthesis, competitive-analysis, stakeholder-summary
problem-framing problem-statement, opportunity-tree, jtbd-canvas
ideation hypothesis, solution-brief
specification prd, user-stories, edge-cases, adr, design-rationale
validation experiment-design, instrumentation-spec, dashboard-requirements
reflection experiment-results, retrospective, lessons-log, pivot-decision
coordination spike-summary, launch-checklist, release-notes, refinement-notes

Suggested Sequence

For a typical feature development, follow this path:

1. competitive-analysis     → Understand market context
2. interview-synthesis      → Gather user insights
3. stakeholder-summary      → Map organizational context
4. problem-statement        → Frame the problem
5. jtbd-canvas              → Understand the job
6. opportunity-tree         → Identify opportunities
7. hypothesis               → Form testable assumptions
8. solution-brief           → Propose solution
9. spike-summary            → Validate feasibility
10. adr                     → Document architecture decisions
11. design-rationale        → Document design decisions
12. prd                     → Specify requirements
13. user-stories            → Break into stories
14. edge-cases              → Cover edge cases
15. instrumentation-spec    → Plan measurement
16. launch-checklist        → Prepare for launch
17. release-notes           → Communicate to users
18. experiment-design       → Plan experiments
19. dashboard-requirements  → Build visibility
20. experiment-results      → Analyze outcomes
21. retrospective           → Reflect as team
22. lessons-log             → Capture learnings
23. refinement-notes        → Plan next iteration
24. pivot-decision          → Decide direction

Not every project needs every skill. Use judgment to select the appropriate subset based on project scope, uncertainty level, and team needs.


Quality Checklist

Before considering this workflow complete, verify:

  • Research covers at least 5 user interviews or equivalent
  • Problem statement is validated with stakeholders
  • Architecture decisions are documented with context and alternatives
  • All user stories have acceptance criteria
  • Experiment design includes guardrail metrics
  • Retrospective action items have owners and deadlines

See Also


Part of PM-Skills — Open source Product Management skills for AI agents