Product-org-os ooda-loop

Document Intelligence

This skill supports three modes: Create, Update, and Find.

Mode Detection

@path/to/ooda-loop.md

Threshold: >=85% auto-proceed | 70-84% state assumption | <70% ask user

Mode Behaviors

CREATE: Walk through all four OODA phases for the given situation, produce a complete cycle document with orientation analysis and action plan.

UPDATE:

1. Read existing OODA analysis (search if path not provided)
2. Preserve prior observations and orientations that remain valid
3. Add new observations, reorient based on new information, update decisions/actions
4. Show diff summary: "New observations: [items]. Reoriented: [sections]. Actions updated: [items]."

FIND:

1. Search paths below for OODA analysis documents
2. Present results: situation, date, cycle number, path
3. Ask: "Update one of these, or create new?"

Search Locations

• decisions/

• strategy/

• product/

• analysis/

Gotchas

• Orientation is the MOST IMPORTANT step — don't rush through it. This is where biases, mental models, and cultural assumptions shape interpretation. Challenge them explicitly.
• OODA is a LOOP, not a linear checklist — the output of Act feeds back into Observe. Always define the feedback mechanism.
• Don't confuse speed with haste — Boyd's insight is about cycle speed relative to competitors, not about skipping analysis.
• Implicit guidance (experienced practitioners skipping Decide/Act) only applies when orientation is deeply internalized — document when this is appropriate vs. when explicit steps are needed.

Vision to Value Phase

Cross-Phase (Rapid Decision Cycle) - OODA applies throughout the Vision to Value flow. It is most valuable during Phase 1 (rapid environmental scanning), Phase 4 (execution tempo), and Phase 6 (learning feedback loops).

Prerequisites: A situation requiring a decision, competitive context, or environmental uncertainty Outputs used by:

/decision-record

/strategic-bet

/experiment-design

/retrospective

Methodology

The OODA Loop

The OODA Loop is a decision-making framework developed by military strategist Colonel John Boyd. The core principle: whoever cycles through OODA fastest gains the initiative and wins. In product strategy, this means faster market sensing, faster interpretation, faster decisions, and faster execution than competitors.

1. Observe (Environmental Scanning)

Gather raw information from the environment:

• Market signals, customer behavior changes, competitive moves
• Internal metrics, usage data, support tickets, churn signals
• Technology shifts, regulatory changes, ecosystem developments
• Team sentiment, organizational friction, process bottlenecks

2. Orient (The Critical Step)

Analyze and synthesize observations into a mental model. This is where competitive advantage lives. Orientation is shaped by:

• Cultural traditions: Industry norms, company culture, "how we've always done it"
• Previous experience: Pattern recognition from past cycles, institutional memory
• New information: Fresh data that challenges existing mental models
• Analysis & synthesis: Breaking apart observations and recombining them into new understanding
• Genetic heritage: Core identity, mission, values that filter interpretation

Boyd's key insight: orientation shapes observation (you see what your mental model allows) AND orientation shapes action (you act based on how you interpret, not on raw data). Challenging orientation biases is the highest-leverage activity.

3. Decide (Select Course of Action)

Based on the oriented mental model, select a course of action:

• Hypothesis: "If we do X, we expect Y because of Z"
• Alternatives considered (minimum 2)
• Reversibility assessment: one-way door vs. two-way door
• Speed vs. accuracy tradeoff: when is "good enough" better than "perfect"?

4. Act (Execute and Feed Back)

Execute the decision and establish the feedback loop:

• Ship the change, launch the experiment, make the move
• Define what to observe in the next cycle (close the loop)
• Set the tempo: how fast should the next cycle begin?
• Communicate the action to the team (shared orientation)

Implicit Guidance and Control

Experienced practitioners develop such strong orientation that they can flow directly from Orient to Act, bypassing explicit Decide. This is analogous to expert intuition — but it requires:

• Deep domain expertise
• Many completed OODA cycles building pattern recognition
• High-trust environment where rapid action is supported

Tempo and Competitive Advantage

Operating INSIDE the competitor's OODA loop means:

• You observe and act before they can orient to your last move
• They are perpetually reacting to a reality you've already moved past
• Their orientation becomes increasingly disconnected from the actual environment

Anti-Patterns

• Analysis paralysis: Stuck in Orient, never reaching Decide
• Fire-ready-aim: Acting without observing (skipping O-O)
• Confirmation bias in Orient: Only seeing data that confirms existing mental model
• Single-loop: Running OODA once as a linear process instead of a continuous cycle
• Individual OODA without shared orientation: Team members cycling at different speeds with different mental models

Output Structure

# OODA Loop Analysis: [Situation / Decision Context]

**Date**: [YYYY-MM-DD]
**Owner**: [Who owns this decision cycle]
**Cycle**: [#1, #2, etc. — increment for recurring analysis]
**Tempo Target**: [How fast should this cycle complete?]
**Context**: [Why this OODA cycle is needed now]

## Observe: Environmental Scan

### External Signals

| # | Observation | Source | Freshness | Reliability |
|---|-------------|--------|-----------|-------------|
| OB1 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB2 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB3 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB4 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB5 | [Signal] | [Source] | [When observed] | High/Med/Low |

### Internal Signals

| # | Observation | Source | Freshness | Reliability |
|---|-------------|--------|-----------|-------------|
| OB6 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB7 | [Signal] | [Source] | [When observed] | High/Med/Low |
| OB8 | [Signal] | [Source] | [When observed] | High/Med/Low |

### Blind Spots
[What are we NOT seeing? What data sources are we missing?]

## Orient: Mental Model Formation

### Current Mental Model
[How do we currently interpret the situation? What is our working hypothesis about what's happening?]

### Orientation Filters (Boyd's Factors)

| Filter | How It Shapes Our View | Challenge |
|--------|----------------------|-----------|
| Cultural traditions | [Industry/company norms influencing interpretation] | [How might this be wrong?] |
| Previous experience | [Past patterns we're matching to] | [Is this situation actually different?] |
| New information | [Data that challenges our existing model] | [Are we weighting this appropriately?] |
| Analysis & synthesis | [How we're combining signals] | [What alternative synthesis exists?] |

### Reoriented Mental Model
[After challenging filters: what is our UPDATED interpretation of the situation?]

### Key Insight
[The single most important realization from orientation — what changed in our understanding?]

## Decide: Course of Action

### Options Considered

| Option | Hypothesis | Reversibility | Speed | Expected Outcome |
|--------|-----------|---------------|-------|-----------------|
| A: [Action] | If we [X], then [Y] because [Z] | One-way / Two-way | [Time] | [Expected result] |
| B: [Action] | If we [X], then [Y] because [Z] | One-way / Two-way | [Time] | [Expected result] |
| C: [Action] | If we [X], then [Y] because [Z] | One-way / Two-way | [Time] | [Expected result] |

### Selected Action
**Option [X]**: [Name]

**Rationale**: [Why this option, given our orientation]
**Speed vs. Accuracy**: [Are we optimizing for tempo or precision in this cycle?]

## Act: Execution Plan

### Immediate Actions

| # | Action | Owner | By When | Done? |
|---|--------|-------|---------|-------|
| 1 | [Action] | [Who] | [When] | [ ] |
| 2 | [Action] | [Who] | [When] | [ ] |
| 3 | [Action] | [Who] | [When] | [ ] |

### Feedback Loop (Close the Cycle)

**What to observe next**: [Specific signals to watch for after acting]
**Next cycle trigger**: [When/what triggers the next OODA iteration]
**Tempo**: [Target time for next complete cycle]
**Shared orientation**: [How will the team be aligned on the updated mental model?]

## Competitive Tempo Assessment

**Our cycle speed**: [Estimated time for this OODA cycle]
**Competitor cycle speed**: [Estimated — are we inside or outside their loop?]
**Tempo advantage**: [Are we gaining or losing initiative?]

## Assumptions

| # | Assumption | Phase | Confidence | If Wrong |
|---|-----------|-------|------------|----------|
| 1 | [Assumption] | O/O/D/A | High/Med/Low | [Impact] |
| 2 | [Assumption] | O/O/D/A | High/Med/Low | [Impact] |

## Next Steps

- [ ] Execute Act items and begin observing results
- [ ] Schedule next OODA cycle based on tempo target
- [ ] Formalize decision via `/decision-record` if high-stakes
- [ ] Feed learnings into `/retrospective` after cycle completes

Instructions

1. Clarify the situation or decision context with the user
2. Check prior context: Run

   /context-recall [topic]

   to find related past decisions or cycles
3. Walk through each OODA phase sequentially — spend the most time on Orient
4. In Orient, explicitly challenge each of Boyd's orientation filters (cultural traditions, experience, new information, analysis/synthesis)
5. Require at least 2 options in the Decide phase — if only one option exists, the decision is already made
6. Define the feedback loop in Act — an OODA cycle without a feedback mechanism is not a loop
7. Assess competitive tempo: are we cycling faster or slower than the competition?
8. Save output as markdown file
9. Offer to formalize high-stakes decisions via

   /decision-record

   or frame strategic bets via

   /strategic-bet

Integration

• Links to

  /decision-record

  (formalize the Decide step for high-stakes choices)
• Links to

  /strategic-bet

  (frame Act as a strategic bet when stakes are high)
• Links to

  /experiment-design

  (structure the Act phase as a formal experiment)
• Links to

  /retrospective

  (evaluate completed OODA cycles for learning)
• Links to

  /competitive-landscape

  (inform Observe phase with competitive data)
• Links to

  /assumption-map

  (track assumptions across OODA phases)
• Links to

  /context-save

  (persist cycle for future reference and cross-cycle learning)