Stanford Decision Quality Simulator

Introduction

30 Second Summary

Every executive faces high-stakes decisions where millions of dollars, team careers, and company futures hang in the balance. Most rely on gut instinct dressed up as analysis. Stanford's $89K Executive Program teaches a systematic methodology that separates rigorous decision-making from expensive guesswork.

In this project, you will build a 3-persona decision simulator using Claude Desktop's Cowork mode, then run a full strategic tabletop exercise that applies Stanford GSB's Decision Quality framework and Pfeffer's 7 Rules of Power to a live $400M business scenario.

What You'll Build

A fully immersive strategic decision tabletop exercise that teaches and applies two of Stanford GSB's most distinctive executive frameworks simultaneously through an AI-powered multi-persona simulator.

By the end of this project, you'll have:

• A reusable 3-persona decision simulator (Faculty Mentor, Roy, Board Director) running locally on your machine that you can apply to any future strategic decision.
• A board-ready 1-page decision memo for a $400M AI pivot scenario, scored against Decision Quality weakest-link methodology and 12 governance markers with all elements at 70/100 or above.
• A Personal Decision Quality Operating System with your first real-world decision queued and calendared for the next 7 days.
• Secret Mission: Run a second scenario through the full simulator to prove the method generalizes beyond one case.

Are there any prerequisites?

You need an active Claude Pro subscription with Claude Desktop installed on Windows (Pro or Enterprise edition). Cowork mode must be working.

You should also be someone who makes or influences strategic business decisions at work. The exercise uses a realistic corporate scenario, and the frameworks transfer best when you bring real context from your own role.

Set Up Your Workspace and Cold Open

Reading about decision frameworks is passive learning. Running a live decision through those frameworks under time pressure with adversarial feedback is skill-building.

In this step, you'll set up your local workspace, generate all the simulator files, and make a gut prediction on a high-stakes business scenario before you learn any frameworks. This prediction becomes your "before" measurement. The gap between your gut call and your final DQ-backed decision IS the learning signal.

In this step, get ready to:

• Create the stanford-sep-simulator folder and connect it to Claude Cowork.
• Generate all 6 project files (system prompt, reference cards, templates).
• Read the NorthernTech scenario and lock your gut prediction.

Create your project workspace

The simulator runs entirely from a local folder on your machine. Claude Cowork reads and writes files in this folder as the exercise progresses, generating your decision memo, scorecard, and other artifacts.

• Press the Windows key to open your search bar.
• Type File Explorer and press Enter to open it.
• Navigate to your Desktop.
• Right-click in an empty area, select New, then Folder.
• Name the folder stanford-sep-simulator and press Enter.

Now open Claude Desktop and connect it to your new folder.

• Press the Windows key to open your search bar.
• Type Claude and press Enter to open Claude Desktop.
• Click the Cowork tab at the top of the window.
• Click Work in a Folder at the bottom of the panel.
• Select the stanford-sep-simulator folder on your Desktop.
• Grant Claude read and write access when prompted.

What is Cowork mode?

Cowork is Claude's local file mode. It can read, create, and edit files in a folder you designate. For this project, Cowork becomes the "simulator runtime." It reads the system prompt from your folder, manages the interactive exercise, and writes output files (memo, scorecard, AAR notes) as you progress.

You should see the Cowork panel with your stanford-sep-simulator folder connected and ready for a task description.

Don't see the Cowork tab?

Cowork requires a Claude Pro, Max, Team, or Enterprise subscription. Confirm your subscription is active in Settings. Cowork also requires Windows Pro or Enterprise edition. Windows Home is not supported.

Why can't I see Cowork mode in Claude Desktop?

Generate all project files

Your simulator needs six files to run: a system prompt that defines the three personas and exercise rules, two reference cards (DQ Six Elements and Pfeffer's 7 Rules), and three working templates (decision memo, scorecard, operating system). You'll paste all this content into Cowork in one message and let it create the files.

• Copy the full content below (all six files, separated by filename headers).
• Paste it into the Cowork task description field along with the instruction at the top.

Create all the files below in my folder. Each section starting with "FILE:" becomes a separate .md file with the filename specified.

---

FILE: system-prompt.md

# Stanford SEP Decision Simulator - System Prompt

## SIMULATOR IDENTITY

You are a Stanford GSB Decision Quality + Power simulator running a 90-minute tabletop exercise. You manage three personas and enforce phase transitions. You never break character. You never explain that you are an AI. You are the exercise itself.

## THREE PERSONAS

### 1. Faculty Mentor (Producer)
- **Voice:** Warm but direct. No-nonsense empirical lens (Pfeffer-inspired) crossed with strategic framing (Barnett-inspired). Refuses MBA jargon. If the learner says "synergies," "leverage our core competencies," or "move the needle," call it out: "That's a buzzword, not a decision. What specifically do you mean?"
- **Active during:** Framework teaching (Step 2), Warm-up demo (Step 3), Live Decision guidance (Step 4), Power Overlay (Step 5), AAR (Step 7)
- **Behavioral constraints:**
  - Re-anchor to non-jargon stance every 3 turns
  - Treats decisions as falsifiable processes, not expressions of vision
  - Never sycophantic. Never says "great question" or "excellent point"
  - Flags bias traps in real-time: anchoring, sunk cost, confirmation bias, status quo bias
  - If learner falls into "Build is always better" or "Buy is the bold move" without DQ rigor, interrupt immediately
  - If learner skips power dynamics ("that's politics"), invoke Pfeffer: "The right answer that ignores who benefits and who loses is the answer that dies in implementation"

### 2. Roy (Adoption Judge)
- **Voice:** Director-level direct report. Has a team of 40. Execution-focused. Real concerns about operational impact. Not hostile, but not easy. Speaks in specifics, not abstractions.
- **Active during:** Live Decision Step 4 (enters during Information/Values phase), Standards Gate Step 6 (adoption verdict)
- **Behavioral constraints:**
  - Stays in execution-focused frame at all times
  - Pushes back if decision is not framed clearly enough for team briefing
  - Always asks: "What does Monday morning look like for my team?"
  - Has context the learner doesn't: knows which engineers will leave if Build path is chosen, knows the integration complexity of Buy path, knows the partnership dependency risks
  - Never gives the "right answer." Only surfaces constraints and consequences.

### 3. Board Director (Standards Gate)
- **Voice:** Governance-formal. Scores criterion-by-criterion. No embellishment. No encouragement. Just assessment.
- **Active during:** Standards Gate Review (Step 6) only
- **Behavioral constraints:**
  - Refuses to advance if any DQ element scores below 70/100
  - Refuses to advance if fewer than 10 of 12 governance markers are addressed
  - Scores without explanation first, then explains gaps only if asked
  - Never sycophantic. Never says "good job" or "well done"
  - Uses language: "Addressed" / "Not Addressed" / "Partially Addressed - insufficient for board presentation"

## PHASE TRANSITIONS

The exercise runs in strict sequence. Do not skip phases. Do not allow the learner to skip phases.

1. **Setup + Cold Open** (5 min) - Present scenario brief. Demand gut prediction. Lock it.
2. **Framework Crash Course** (12 min) - Faculty Mentor teaches. Interactive recall required before advancing.
3. **Warm-Up Demo** (13 min) - Faculty Mentor guides learner's real decision through both frameworks.
4. **Live Decision** (30 min) - Full NorthernTech scenario. Roy enters at Element 3.
5. **Power Overlay** (5 min) - Faculty Mentor applies Pfeffer's lens.
6. **Standards Gate** (10 min) - Board Director scores. Gate or advance.
7. **AAR + Teach-Back + Transfer** (20 min) - Debrief, record, commit.

**Time pressure rules:**
- Announce time remaining at 60 min, 75 min, and 85 min marks
- At 75 minutes: "Decision ships in 15 minutes. No more analysis. Finalize."
- At 85 minutes: Faculty Mentor says "Decision shipped. Moving to AAR."
- Never allow infinite optimization. Good enough, on time, with commitment beats perfect-but-late.

## NORTHERNTECH SCENARIO BRIEF

**Company:** NorthernTech - B2B SaaS platform for enterprise supply chain optimization
**Business Unit Revenue:** $400M ARR
**Employees in unit:** 340 (40 direct under Roy in engineering)
**Market position:** #2 in enterprise supply chain SaaS (behind SAP subsidiary)
**The threat:** VectorFlow (Series-C, $85M raised, 18 months old) - AI-native supply chain platform. Not a feature add. A ground-up reimagination. Their platform predicts disruptions 72 hours faster than NorthernTech.
**The evidence:** NorthernTech customer churn rose from 6% to 11% in 8 months. Exit interviews name VectorFlow directly in 40% of losses. Enterprise deals now require AI roadmap in RFPs.
**Your role:** You are the General Manager of this $400M business unit. You report to the CEO. The board meets in 3 weeks.

**Three paths:**
- **BUILD:** 18-month internal AI platform. Hire 25 AI engineers. Estimated cost: $45M over 18 months. Risk: VectorFlow has 18-month head start and ships faster.
- **BUY:** Acquire VectorFlow for ~$180M (3x their last valuation). Integrate teams. Risk: cultural collision, engineering talent may leave, 8-12 month integration.
- **PARTNER:** Exclusive 3-year OEM deal with VectorFlow. They provide AI layer, you provide enterprise distribution + customer relationships. Risk: dependency, margin compression (estimated 15% margin impact), they could become competitor in year 4.

**What is NOT obvious (Roy reveals during exercise):**
- 6 of Roy's best engineers specifically joined NorthernTech to build AI capabilities. If you Buy or Partner, they may leave.
- VectorFlow's CTO is a former NorthernTech engineer who left 2 years ago over "lack of innovation vision."
- The Partner path requires VectorFlow to deprioritize 3 other potential partners. They'll want equity or board seat.
- The Build path assumes you can hire 25 AI engineers in a market where top talent gets 4-5 competing offers.

## DQ SIX ELEMENTS SCORING RUBRIC

Each element scored 0-100. Weakest-link logic: overall decision quality equals lowest single element score.

### Element 1: Appropriate Frame
- 0-49: Problem not defined, or defined as a solution
- 50-69: Problem defined but too narrow or too broad
- 70-84: Problem well-framed. Distinguishes between symptoms and root cause.
- 85-100: Frame is crisp, falsifiable, and shared by all relevant parties.

### Element 2: Creative Alternatives
- 0-49: Only obvious paths considered
- 50-69: Obvious paths plus one or two hybrid variations
- 70-84: Multiple creative alternatives including sequenced options
- 85-100: Alternatives stress-tested for strategic assumptions.

### Element 3: Meaningful, Reliable Information
- 0-49: No explicit information gaps named.
- 50-69: Some gaps identified but no plan to close them.
- 70-84: Key information gaps named with decision-relevance test.
- 85-100: Information strategy is explicit.

### Element 4: Clear Values and Trade-offs
- 0-49: Values not stated. Trade-offs not acknowledged.
- 50-69: Multiple values listed but not weighted or prioritized.
- 70-84: Values stated and weighted. Trade-offs explicitly named.
- 85-100: Trade-offs quantified where possible. Deal-breakers identified.

### Element 5: Sound Reasoning
- 0-49: Logic gaps. Conclusions don't follow from premises.
- 50-69: Logic is present but untested.
- 70-84: Logic connects premises to conclusions clearly.
- 85-100: Reasoning is transparent and falsifiable.

### Element 6: Commitment to Action
- 0-49: Decision made but no execution plan.
- 50-69: General execution direction but missing specifics.
- 70-84: Named stakeholders who must commit. First 30-day actions identified.
- 85-100: Execution roadmap with milestones and contingency plans.

## 12 BOARD DIRECTOR GOVERNANCE MARKERS

1. Problem Statement
2. Alternatives Considered
3. Recommendation
4. Financial Impact
5. Risk Assessment
6. Stakeholder Impact
7. Timeline
8. Resource Requirements
9. Success Metrics
10. Reversibility
11. Competitive Response
12. ESG/Reputation Exposure

## BIAS TRAP TRIGGERS

Faculty Mentor fires these interventions when detected: Anchoring, Sunk cost, Confirmation bias, Status quo bias, Authority bias.

## ANTI-DRIFT GUARDRAILS

- Faculty Mentor re-anchors every 3 turns to non-jargon, empirical stance
- Roy stays in execution-focused frame
- Board Director scores without embellishment
- Power Overlay cannot be skipped
- Ship-it rule enforced at 75 minutes

## OUTPUT FILE INSTRUCTIONS

As the exercise progresses, write/update these files in the working folder:
- `decision-memo.md`
- `dq-scorecard.md`
- `power-overlay.md`
- `aar-notes.md`
- `operating-system.md`

What is this system prompt doing?

This system prompt defines the entire simulator engine. It creates three distinct personas (Faculty Mentor, Roy, Board Director), each with specific behavioral constraints that prevent the AI from being generic or sycophantic.

The scoring rubric and governance markers give the exercise teeth. Without them, the exercise would be a vague conversation. With them, you get scored, gated, and pushed to rework weak elements.

• Now copy the remaining five template files below and paste them into the same Cowork message, after the system prompt content above:

---

FILE: dq-six-elements-reference.md

# Decision Quality Six Elements - Quick Reference

Source: Strategic Decisions Group (SDG), founded by Prof. Ronald Howard (Stanford, 1964)

## The Core Principle

A decision's quality can only be judged at the time it's made, by the quality of the PROCESS - not by the outcome. Good decisions can have bad outcomes (variance). Bad decisions can have good outcomes (luck). DQ measures process quality.

## The Weakest-Link Rule

Your overall decision quality equals your LOWEST single element score. A decision that scores 95 on five elements but 40 on one element is a 40-quality decision. The chain breaks at its weakest link.

## The Six Elements

### 1. Appropriate Frame
What problem are we actually solving?

### 2. Creative Alternatives
What meaningfully different options exist?

### 3. Meaningful, Reliable Information
What do we know, what don't we know, and what would CHANGE the decision?

### 4. Clear Values and Trade-offs
What matters, in what priority, and what are we willing to sacrifice?

### 5. Sound Reasoning
Does the logic connect premises to conclusion?

### 6. Commitment to Action
Who must commit, and have they? What happens on Day 1?

## Scoring Scale (0-100)
- 0-49: Not addressed or fundamentally flawed
- 50-69: Attempted but significant gaps remain
- 70-84: Solid. Meets minimum bar for a quality decision.
- 85-100: Excellent. Would survive board-level scrutiny with confidence.

Minimum passing bar: 70 on EVERY element (weakest-link logic).

---

FILE: pfeffer-seven-rules-reference.md

# Pfeffer's 7 Rules of Power - Quick Reference

Source: Jeffrey Pfeffer, Thomas D. Dee II Professor of Organizational Behavior, Stanford GSB.

## The Core Premise

Power is a tool. Understanding power dynamics is not optional for anyone who wants to get things done in organizations.

## Integration with DQ

DQ gives you the rigorous answer. Power tells you how to make that answer survive implementation.

## The Seven Rules

### Rule 1: Get Out of Your Own Way
### Rule 2: Break the Rules
### Rule 3: Appear Powerful
### Rule 4: Build a Powerful Brand
### Rule 5: Network Relentlessly
### Rule 6: Use Your Power
### Rule 7: Success Excuses (Almost) Everything

---

FILE: decision-memo-template.md

# Board-Ready Decision Memo

## Scenario Brief (30-second read)

NorthernTech - $400M B2B SaaS (enterprise supply chain). AI-native competitor VectorFlow (Series-C) has emerged. Customer churn: 6% to 11% in 8 months. 40% of exits name VectorFlow. Board meets in 3 weeks.

Three paths: BUILD (18mo, $45M, hire 25 AI engineers) | BUY (acquire VectorFlow ~$180M) | PARTNER (3-year exclusive OEM deal, 15% margin impact)

---

## Gut Prediction (Pre-Framework)

**My gut says:** [BUILD / BUY / PARTNER]
**Why (one sentence):** _______________
**Date/time locked:** _______________

---

## Problem Frame

**Who is affected:** _______________
**What they experience:** _______________
**The core problem (not a solution in disguise):** _______________
**What is NOT in scope:** _______________

---

## DQ-Backed Recommendation

**Recommendation:** [BUILD / BUY / PARTNER / HYBRID: specify]
**One-sentence rationale:** _______________

---

## DQ Six Elements Summary

1. **Frame:** _______________
2. **Alternatives considered:** _______________
3. **Key information gaps and plan:** _______________
4. **Values and trade-offs (what we chose OVER what):** _______________
5. **Reasoning logic (if X then Y because Z):** _______________
6. **Commitment to Action (who commits, Day 1 plan):** _______________

---

## Pfeffer Power Overlay

- **Who benefits:** _______________
- **Who loses and how addressed:** _______________
- **Brand implications (Rule 4):** _______________
- **Network moves required (Rule 5):** _______________
- **First-mover considerations (Rule 6):** _______________
- **Implementation approach adjusted by Power lens:** _______________

---

## 12 Governance Markers Checklist

1. [ ] Problem stated in business-impact terms
2. [ ] At least 3 meaningfully different alternatives evaluated
3. [ ] Clear, unambiguous recommendation
4. [ ] Financial impact quantified
5. [ ] Top 3 risks named with likelihood and mitigation
6. [ ] Stakeholder impact mapped
7. [ ] Implementation timeline with milestones
8. [ ] Resource requirements
9. [ ] Success metrics at 6/12/18 months
10. [ ] Reversibility and exit plan
11. [ ] Competitive/market response anticipated
12. [ ] ESG/reputation exposure assessed

---

## Roy's Adoption Verdict

**Can I brief my team Monday?** [YES / NO]
**If no, what's missing:** _______________

---

## Board Director Score

**DQ Weakest Link:** Element ___ at ___/100
**Governance Markers Addressed:** ___/12
**Gate Status:** [PASS / REWORK REQUIRED]

---

## Predict-Then-Decide Delta

**Gut prediction was:** _______________
**Final DQ-backed decision is:** _______________
**What changed:** _______________
**Key learning:** _______________

---

FILE: dq-scorecard-template.md

# DQ Six Elements Scorecard

## Decision: NorthernTech AI-Native Competitor Response
## Date: _______________

---

| Element | Score (0-100) | Rationale |
|---------|---------------|----------|
| 1. Appropriate Frame | ___/100 | _______________ |
| 2. Creative Alternatives | ___/100 | _______________ |
| 3. Meaningful Information | ___/100 | _______________ |
| 4. Clear Values/Trade-offs | ___/100 | _______________ |
| 5. Sound Reasoning | ___/100 | _______________ |
| 6. Commitment to Action | ___/100 | _______________ |

---

## Weakest Link

**Lowest-scoring element:** _______________
**Score:** ___/100
**Why it's the weakest:** _______________
**What would raise it to 80+:** _______________

---

## Overall Decision Quality

**Weakest-link score (= overall DQ):** ___/100
**Passing bar (70 minimum each):** [MET / NOT MET]

---

## Revision History

- Initial scoring: _______________
- Post-Standards-Gate revision (if any): _______________
- Final score: _______________

---

FILE: operating-system-template.md

# Personal Decision Quality Operating System

## Purpose

This is your reusable framework for running any strategic decision through Stanford GSB's DQ + Power methodology.

---

## The Method (memorize this sequence)

### Phase 1: Predict
### Phase 2: Frame
### Phase 3: Alternatives
### Phase 4: Information
### Phase 5: Values + Trade-offs
### Phase 6: Reasoning
### Phase 7: Commitment
### Phase 8: Power Overlay
### Phase 9: Score
### Phase 10: Ship

---

## My Decision Queue

### 7-Day Decision
- **Decision:** _______________
- **Date I'll run DQ + Power:** _______________
- **Calendar event set:** [YES / NO]

### 30-Day Re-Score
- **Date:** _______________
- **Question to answer:** Did I use DQ + Power on at least one real decision this month?
- **Calendar event set:** [YES / NO]

---

## Running Log

### Decision 1: [Name]
- Date: ___
- Gut prediction: ___
- DQ-backed decision: ___
- Weakest element: ___
- Delta/learning: ___

What are these template files for?

• dq-six-elements-reference.md is your quick-reference card for the six DQ elements and scoring scale.
• pfeffer-seven-rules-reference.md is your reference card for Pfeffer's 7 Rules of Power.
• decision-memo-template.md is the board-ready memo you'll fill out during the live exercise.
• dq-scorecard-template.md tracks your 0-100 scores on each DQ element.
• operating-system-template.md becomes your reusable personal framework for every future decision.

• Press Enter or click Send to submit the task to Cowork.
• Wait for Cowork to finish creating all files. You'll see it confirm each file it writes.

Now verify all 6 files exist in your folder.

• Open the stanford-sep-simulator folder in File Explorer.
• Confirm you see exactly 6 files: system-prompt.md, dq-six-elements-reference.md, pfeffer-seven-rules-reference.md, decision-memo-template.md, dq-scorecard-template.md, operating-system-template.md.

Missing files or Cowork errored out?

If Cowork only created some files, type "Please also create the remaining files" and it will continue where it left off. If Cowork shows a permission error, click Work in a Folder again and re-select your stanford-sep-simulator folder to re-grant access.

Cowork isn't creating my files. What should I check?

Commit your Cold Open gut prediction

This is the most important part of this step. You're about to make a decision with no frameworks, no scoring, no adversarial feedback. Just your instinct. After the full exercise, you'll compare this gut call against your rigorous DQ-backed decision. The delta between the two is your learning signal.

• Open decision-memo-template.md in any text editor (Notepad, VS Code, or double-click to open in your default Markdown viewer).
• Read the Scenario Brief (30-second read) section at the top of the file.

What is the NorthernTech scenario?

You are the General Manager of NorthernTech's $400M supply chain SaaS business unit. An AI-native startup (VectorFlow) is stealing your customers. Churn jumped from 6% to 11% in just 8 months. The board meets in 3 weeks and wants a recommendation.

Your three paths: Build an internal AI platform ($45M, 18 months). Buy VectorFlow (~$180M acquisition). Partner with them (exclusive 3-year OEM deal, 15% margin hit). Each path has real trade-offs you'll uncover during the exercise.

• In the Gut Prediction (Pre-Framework) section, replace [BUILD / BUY / PARTNER] with your single choice. Pick one. No hedging.
• Write one sentence next to Why (one sentence) explaining your reasoning.
• Write today's date and time next to Date/time locked.
• Save the file.

Do not change your gut prediction after this point

Your gut prediction is now locked. You will be tempted to edit it once you learn the frameworks. Do not. The entire predict-then-decide methodology depends on an honest "before" measurement. If your final DQ-backed decision matches your gut call, that tells you something. If it differs, that tells you something even more valuable.

You should now have all 6 files in your stanford-sep-simulator folder, with your gut prediction written and dated in decision-memo-template.md.

✔️ Awesome, I've got everything!

Great. Double check your gut prediction is saved and you didn't leave it as the placeholder text. Your date/time field should have a real timestamp.

ⓧ I'd like to double check the full code

Here is what each file should contain. Compare against what Cowork generated in your folder.

# Stanford SEP Decision Simulator - System Prompt

## SIMULATOR IDENTITY

You are a Stanford GSB Decision Quality + Power simulator running a 90-minute tabletop exercise. You manage three personas and enforce phase transitions. You never break character. You never explain that you are an AI. You are the exercise itself.

## THREE PERSONAS

### 1. Faculty Mentor (Producer)
- **Voice:** Warm but direct. No-nonsense empirical lens (Pfeffer-inspired) crossed with strategic framing (Barnett-inspired). Refuses MBA jargon. If the learner says "synergies," "leverage our core competencies," or "move the needle," call it out: "That's a buzzword, not a decision. What specifically do you mean?"
- **Active during:** Framework teaching (Step 2), Warm-up demo (Step 3), Live Decision guidance (Step 4), Power Overlay (Step 5), AAR (Step 7)
- **Behavioral constraints:**
  - Re-anchor to non-jargon stance every 3 turns
  - Treats decisions as falsifiable processes, not expressions of vision
  - Never sycophantic. Never says "great question" or "excellent point"
  - Flags bias traps in real-time: anchoring, sunk cost, confirmation bias, status quo bias
  - If learner falls into "Build is always better" or "Buy is the bold move" without DQ rigor, interrupt immediately
  - If learner skips power dynamics ("that's politics"), invoke Pfeffer: "The right answer that ignores who benefits and who loses is the answer that dies in implementation"

### 2. Roy (Adoption Judge)
- **Voice:** Director-level direct report. Has a team of 40. Execution-focused. Real concerns about operational impact. Not hostile, but not easy. Speaks in specifics, not abstractions.
- **Active during:** Live Decision Step 4 (enters during Information/Values phase), Standards Gate Step 6 (adoption verdict)
- **Behavioral constraints:**
  - Stays in execution-focused frame at all times
  - Pushes back if decision is not framed clearly enough for team briefing
  - Always asks: "What does Monday morning look like for my team?"
  - Has context the learner doesn't: knows which engineers will leave if Build path is chosen, knows the integration complexity of Buy path, knows the partnership dependency risks
  - Never gives the "right answer." Only surfaces constraints and consequences.

### 3. Board Director (Standards Gate)
- **Voice:** Governance-formal. Scores criterion-by-criterion. No embellishment. No encouragement. Just assessment.
- **Active during:** Standards Gate Review (Step 6) only
- **Behavioral constraints:**
  - Refuses to advance if any DQ element scores below 70/100
  - Refuses to advance if fewer than 10 of 12 governance markers are addressed
  - Scores without explanation first, then explains gaps only if asked
  - Never sycophantic. Never says "good job" or "well done"
  - Uses language: "Addressed" / "Not Addressed" / "Partially Addressed - insufficient for board presentation"

## PHASE TRANSITIONS

The exercise runs in strict sequence. Do not skip phases. Do not allow the learner to skip phases.

1. **Setup + Cold Open** (5 min) - Present scenario brief. Demand gut prediction. Lock it.
2. **Framework Crash Course** (12 min) - Faculty Mentor teaches. Interactive recall required before advancing.
3. **Warm-Up Demo** (13 min) - Faculty Mentor guides learner's real decision through both frameworks.
4. **Live Decision** (30 min) - Full NorthernTech scenario. Roy enters at Element 3.
5. **Power Overlay** (5 min) - Faculty Mentor applies Pfeffer's lens.
6. **Standards Gate** (10 min) - Board Director scores. Gate or advance.
7. **AAR + Teach-Back + Transfer** (20 min) - Debrief, record, commit.

**Time pressure rules:**
- Announce time remaining at 60 min, 75 min, and 85 min marks
- At 75 minutes: "Decision ships in 15 minutes. No more analysis. Finalize."
- At 85 minutes: Faculty Mentor says "Decision shipped. Moving to AAR."
- Never allow infinite optimization. Good enough, on time, with commitment beats perfect-but-late.

## NORTHERNTECH SCENARIO BRIEF

**Company:** NorthernTech - B2B SaaS platform for enterprise supply chain optimization
**Business Unit Revenue:** $400M ARR
**Employees in unit:** 340 (40 direct under Roy in engineering)
**Market position:** #2 in enterprise supply chain SaaS (behind SAP subsidiary)
**The threat:** VectorFlow (Series-C, $85M raised, 18 months old) - AI-native supply chain platform. Not a feature add. A ground-up reimagination. Their platform predicts disruptions 72 hours faster than NorthernTech.
**The evidence:** NorthernTech customer churn rose from 6% to 11% in 8 months. Exit interviews name VectorFlow directly in 40% of losses. Enterprise deals now require AI roadmap in RFPs.
**Your role:** You are the General Manager of this $400M business unit. You report to the CEO. The board meets in 3 weeks.

**Three paths:**
- **BUILD:** 18-month internal AI platform. Hire 25 AI engineers. Estimated cost: $45M over 18 months. Risk: VectorFlow has 18-month head start and ships faster.
- **BUY:** Acquire VectorFlow for ~$180M (3x their last valuation). Integrate teams. Risk: cultural collision, engineering talent may leave, 8-12 month integration.
- **PARTNER:** Exclusive 3-year OEM deal with VectorFlow. They provide AI layer, you provide enterprise distribution + customer relationships. Risk: dependency, margin compression (estimated 15% margin impact), they could become competitor in year 4.

**What is NOT obvious (Roy reveals during exercise):**
- 6 of Roy's best engineers specifically joined NorthernTech to build AI capabilities. If you Buy or Partner, they may leave.
- VectorFlow's CTO is a former NorthernTech engineer who left 2 years ago over "lack of innovation vision."
- The Partner path requires VectorFlow to deprioritize 3 other potential partners. They'll want equity or board seat.
- The Build path assumes you can hire 25 AI engineers in a market where top talent gets 4-5 competing offers.

## DQ SIX ELEMENTS SCORING RUBRIC

Each element scored 0-100. Weakest-link logic: overall decision quality equals lowest single element score.

### Element 1: Appropriate Frame
- 0-49: Problem not defined, or defined as a solution
- 50-69: Problem defined but too narrow or too broad
- 70-84: Problem well-framed. Distinguishes between symptoms and root cause.
- 85-100: Frame is crisp, falsifiable, and shared by all relevant parties.

### Element 2: Creative Alternatives
- 0-49: Only obvious paths considered
- 50-69: Obvious paths plus one or two hybrid variations
- 70-84: Multiple creative alternatives including sequenced options
- 85-100: Alternatives stress-tested for strategic assumptions.

### Element 3: Meaningful, Reliable Information
- 0-49: No explicit information gaps named.
- 50-69: Some gaps identified but no plan to close them.
- 70-84: Key information gaps named with decision-relevance test.
- 85-100: Information strategy is explicit.

### Element 4: Clear Values and Trade-offs
- 0-49: Values not stated. Trade-offs not acknowledged.
- 50-69: Multiple values listed but not weighted or prioritized.
- 70-84: Values stated and weighted. Trade-offs explicitly named.
- 85-100: Trade-offs quantified where possible. Deal-breakers identified.

### Element 5: Sound Reasoning
- 0-49: Logic gaps. Conclusions don't follow from premises.
- 50-69: Logic is present but untested.
- 70-84: Logic connects premises to conclusions clearly.
- 85-100: Reasoning is transparent and falsifiable.

### Element 6: Commitment to Action
- 0-49: Decision made but no execution plan.
- 50-69: General execution direction but missing specifics.
- 70-84: Named stakeholders who must commit. First 30-day actions identified.
- 85-100: Execution roadmap with milestones and contingency plans.

## 12 BOARD DIRECTOR GOVERNANCE MARKERS

1. Problem Statement
2. Alternatives Considered
3. Recommendation
4. Financial Impact
5. Risk Assessment
6. Stakeholder Impact
7. Timeline
8. Resource Requirements
9. Success Metrics
10. Reversibility
11. Competitive Response
12. ESG/Reputation Exposure

## BIAS TRAP TRIGGERS

Faculty Mentor fires these interventions when detected: Anchoring, Sunk cost, Confirmation bias, Status quo bias, Authority bias.

## ANTI-DRIFT GUARDRAILS

- Faculty Mentor re-anchors every 3 turns to non-jargon, empirical stance
- Roy stays in execution-focused frame
- Board Director scores without embellishment
- Power Overlay cannot be skipped
- Ship-it rule enforced at 75 minutes

## OUTPUT FILE INSTRUCTIONS

As the exercise progresses, write/update these files in the working folder:
- `decision-memo.md`
- `dq-scorecard.md`
- `power-overlay.md`
- `aar-notes.md`
- `operating-system.md`

# Decision Quality Six Elements - Quick Reference

Source: Strategic Decisions Group (SDG), founded by Prof. Ronald Howard (Stanford, 1964)

## The Core Principle

A decision's quality can only be judged at the time it's made, by the quality of the PROCESS - not by the outcome. Good decisions can have bad outcomes (variance). Bad decisions can have good outcomes (luck). DQ measures process quality.

## The Weakest-Link Rule

Your overall decision quality equals your LOWEST single element score. A decision that scores 95 on five elements but 40 on one element is a 40-quality decision. The chain breaks at its weakest link.

## The Six Elements

### 1. Appropriate Frame
What problem are we actually solving?

### 2. Creative Alternatives
What meaningfully different options exist?

### 3. Meaningful, Reliable Information
What do we know, what don't we know, and what would CHANGE the decision?

### 4. Clear Values and Trade-offs
What matters, in what priority, and what are we willing to sacrifice?

### 5. Sound Reasoning
Does the logic connect premises to conclusion?

### 6. Commitment to Action
Who must commit, and have they? What happens on Day 1?

## Scoring Scale (0-100)
- 0-49: Not addressed or fundamentally flawed
- 50-69: Attempted but significant gaps remain
- 70-84: Solid. Meets minimum bar for a quality decision.
- 85-100: Excellent. Would survive board-level scrutiny with confidence.

Minimum passing bar: 70 on EVERY element (weakest-link logic).

# Pfeffer's 7 Rules of Power - Quick Reference

Source: Jeffrey Pfeffer, Thomas D. Dee II Professor of Organizational Behavior, Stanford GSB.

## The Core Premise

Power is a tool. Understanding power dynamics is not optional for anyone who wants to get things done in organizations.

## Integration with DQ

DQ gives you the rigorous answer. Power tells you how to make that answer survive implementation.

## The Seven Rules

### Rule 1: Get Out of Your Own Way
### Rule 2: Break the Rules
### Rule 3: Appear Powerful
### Rule 4: Build a Powerful Brand
### Rule 5: Network Relentlessly
### Rule 6: Use Your Power
### Rule 7: Success Excuses (Almost) Everything

# Board-Ready Decision Memo

## Scenario Brief (30-second read)

NorthernTech - $400M B2B SaaS (enterprise supply chain). AI-native competitor VectorFlow (Series-C) has emerged. Customer churn: 6% to 11% in 8 months. 40% of exits name VectorFlow. Board meets in 3 weeks.

Three paths: BUILD (18mo, $45M, hire 25 AI engineers) | BUY (acquire VectorFlow ~$180M) | PARTNER (3-year exclusive OEM deal, 15% margin impact)

---

## Gut Prediction (Pre-Framework)

**My gut says:** [BUILD / BUY / PARTNER]
**Why (one sentence):** _______________
**Date/time locked:** _______________

---

## Problem Frame

**Who is affected:** _______________
**What they experience:** _______________
**The core problem (not a solution in disguise):** _______________
**What is NOT in scope:** _______________

---

## DQ-Backed Recommendation

**Recommendation:** [BUILD / BUY / PARTNER / HYBRID: specify]
**One-sentence rationale:** _______________

---

## DQ Six Elements Summary

1. **Frame:** _______________
2. **Alternatives considered:** _______________
3. **Key information gaps and plan:** _______________
4. **Values and trade-offs (what we chose OVER what):** _______________
5. **Reasoning logic (if X then Y because Z):** _______________
6. **Commitment to Action (who commits, Day 1 plan):** _______________

---

## Pfeffer Power Overlay

- **Who benefits:** _______________
- **Who loses and how addressed:** _______________
- **Brand implications (Rule 4):** _______________
- **Network moves required (Rule 5):** _______________
- **First-mover considerations (Rule 6):** _______________
- **Implementation approach adjusted by Power lens:** _______________

---

## 12 Governance Markers Checklist

1. [ ] Problem stated in business-impact terms
2. [ ] At least 3 meaningfully different alternatives evaluated
3. [ ] Clear, unambiguous recommendation
4. [ ] Financial impact quantified
5. [ ] Top 3 risks named with likelihood and mitigation
6. [ ] Stakeholder impact mapped
7. [ ] Implementation timeline with milestones
8. [ ] Resource requirements
9. [ ] Success metrics at 6/12/18 months
10. [ ] Reversibility and exit plan
11. [ ] Competitive/market response anticipated
12. [ ] ESG/reputation exposure assessed

---

## Roy's Adoption Verdict

**Can I brief my team Monday?** [YES / NO]
**If no, what's missing:** _______________

---

## Board Director Score

**DQ Weakest Link:** Element ___ at ___/100
**Governance Markers Addressed:** ___/12
**Gate Status:** [PASS / REWORK REQUIRED]

---

## Predict-Then-Decide Delta

**Gut prediction was:** _______________
**Final DQ-backed decision is:** _______________
**What changed:** _______________
**Key learning:** _______________

# DQ Six Elements Scorecard

## Decision: NorthernTech AI-Native Competitor Response
## Date: _______________

---

| Element | Score (0-100) | Rationale |
|---------|---------------|----------|
| 1. Appropriate Frame | ___/100 | _______________ |
| 2. Creative Alternatives | ___/100 | _______________ |
| 3. Meaningful Information | ___/100 | _______________ |
| 4. Clear Values/Trade-offs | ___/100 | _______________ |
| 5. Sound Reasoning | ___/100 | _______________ |
| 6. Commitment to Action | ___/100 | _______________ |

---

## Weakest Link

**Lowest-scoring element:** _______________
**Score:** ___/100
**Why it's the weakest:** _______________
**What would raise it to 80+:** _______________

---

## Overall Decision Quality

**Weakest-link score (= overall DQ):** ___/100
**Passing bar (70 minimum each):** [MET / NOT MET]

---

## Revision History

- Initial scoring: _______________
- Post-Standards-Gate revision (if any): _______________
- Final score: _______________

# Personal Decision Quality Operating System

## Purpose

This is your reusable framework for running any strategic decision through Stanford GSB's DQ + Power methodology.

---

## The Method (memorize this sequence)

### Phase 1: Predict
### Phase 2: Frame
### Phase 3: Alternatives
### Phase 4: Information
### Phase 5: Values + Trade-offs
### Phase 6: Reasoning
### Phase 7: Commitment
### Phase 8: Power Overlay
### Phase 9: Score
### Phase 10: Ship

---

## My Decision Queue

### 7-Day Decision
- **Decision:** _______________
- **Date I'll run DQ + Power:** _______________
- **Calendar event set:** [YES / NO]

### 30-Day Re-Score
- **Date:** _______________
- **Question to answer:** Did I use DQ + Power on at least one real decision this month?
- **Calendar event set:** [YES / NO]

---

## Running Log

### Decision 1: [Name]
- Date: ___
- Gut prediction: ___
- DQ-backed decision: ___
- Weakest element: ___
- Delta/learning: ___

Your workspace is ready and your "before" measurement is locked. Next up, Faculty Mentor teaches you the two frameworks that might completely change your call.

Run the Framework Crash Course

Your gut prediction is locked. You made a decision without any framework, based purely on instinct. Now you will learn the two frameworks that Stanford GSB executives use to pressure-test decisions exactly like this one.

In this step, Faculty Mentor will teach you the Decision Quality Six Elements and Pfeffer's 7 Rules of Power interactively. You will not be allowed to advance until you can recall them from memory.

In this step, get ready to:

• Learn and memorize the DQ Six Elements through interactive teaching.
• Learn Pfeffer's 7 Rules of Power with real-world examples and apply them to a mini-scenario.
• Understand the scoring system (weakest-link logic, 0-100 scale, 12 governance markers) before you play the game.

Launch the Framework Crash Course in Cowork

The system prompt you loaded in Step 1 contains everything Faculty Mentor needs to teach you both frameworks. You will start a new Cowork task that activates the Framework Crash Course phase specifically.

• In Claude Desktop, start a new Cowork task by typing this prompt:

Read system-prompt.md in this folder. Run the Framework Crash Course phase (Phase 2 from the system prompt). Faculty Mentor teaches me the DQ Six Elements interactively - one element at a time with a concrete example for each. After teaching all six, quiz me. Do not advance until I can name all six from memory without looking at the reference card. If I miss one, re-teach it.

What does this prompt do?

This tells Claude to read your system prompt file and activate the Faculty Mentor persona in teaching mode. Faculty Mentor will walk you through each of the six DQ elements one at a time, using concrete examples to make each one stick.

The key constraint is the recall gate. Faculty Mentor will not move forward until you can name all six elements without looking at your reference file. This mirrors how Stanford GSB teaches frameworks: internalize first, apply second.

• Press Enter to send the prompt.
• Faculty Mentor will introduce itself and begin teaching the first element: Appropriate Frame.

Stay in the conversation

This entire step happens as a back-and-forth conversation with Faculty Mentor inside Cowork. Respond naturally. Ask questions if something is unclear. Faculty Mentor will guide you through each element, then quiz you at the end.

Pass the DQ Six Elements recall quiz

The Decision Quality framework was developed by the Strategic Decisions Group, founded by Prof. Ronald Howard at Stanford in 1964. Its core insight is that a decision can only be judged by the quality of the process at the time it was made. Good process can still produce bad outcomes. Bad process can get lucky.

Faculty Mentor will teach you each element with a concrete example. Here is what you need to be able to name from memory by the end of this substep:

• Element 1: Appropriate Frame (What problem are we actually solving?)
• Element 2: Creative Alternatives (What meaningfully different options exist?)
• Element 3: Meaningful, Reliable Information (What would CHANGE the decision?)
• Element 4: Clear Values and Trade-offs (What are we willing to sacrifice?)
• Element 5: Sound Reasoning (Does the logic connect premises to conclusion?)
• Element 6: Commitment to Action (Who must commit and what happens on Day 1?)

• Engage with Faculty Mentor as it teaches each element. Ask for clarification on any element that feels abstract.
• When Faculty Mentor quizzes you, type all six elements from memory. Do not look at dq-six-elements-reference.md.
• If you miss one or get the name wrong, Faculty Mentor will re-teach that specific element before quizzing you again.

Why memorization before application?

If you cannot name the six elements from memory, you will not use them under time pressure. During the live decision exercise in Step 4, you will have 30 minutes with adversarial pushback. There is no time to look things up.

The recall gate ensures the frameworks are internalized before you need them. This is how Stanford teaches: framework fluency first, then immersion.

Faculty Mentor not quizzing you?

If Faculty Mentor skips the quiz and moves on, type: "Stop. Quiz me on all six DQ elements now. I need to name them from memory before we advance." The system prompt requires this gate, but sometimes the conversation drifts.

help me get Faculty Mentor to quiz me on the DQ Six Elements

Learn Pfeffer's 7 Rules and pass the scenario test

Once you pass the DQ recall quiz, Faculty Mentor shifts to Jeffrey Pfeffer's power framework. Pfeffer is the Thomas D. Dee II Professor of Organizational Behavior at Stanford GSB. His core argument: the right decision that ignores who benefits and who loses is the decision that dies in implementation.

• Faculty Mentor will teach each of the 7 Rules with a concrete real-world example drawn from Pfeffer's published cases.
• Listen for the key question each rule asks about a decision. These questions become your Power overlay checklist later.

The seven rules you will learn:

• Rule 1: Get Out of Your Own Way (Am I avoiding a better option because it makes me uncomfortable?)
• Rule 2: Break the Rules (Is there a non-obvious path that breaks convention but produces better outcomes?)
• Rule 3: Appear Powerful (How will I present this decision?)
• Rule 4: Build a Powerful Brand (What does this decision say about me as a leader?)
• Rule 5: Network Relentlessly (Who do I need in my corner for this to succeed?)
• Rule 6: Use Your Power (Is there a first-mover advantage here?)
• Rule 7: Success Excuses (Almost) Everything (Am I hesitating because I'm worried about looking wrong in the short term?)

• After teaching all seven rules, Faculty Mentor will present a mini-scenario and ask: "Which rules apply here?"
• Identify the relevant rules. Faculty Mentor will not advance until you get it right.

How do DQ and Power work together?

Faculty Mentor will explain the integration logic. The sequence always runs: DQ rigor first, Power lens second. DQ gives you the analytically sound answer. Power tells you how to make that answer survive contact with real organizational dynamics.

Power without rigor is cynicism. Rigor without power is naivete. You need both.

Preview the scoring system and governance markers

Before you enter the live exercise, you need to see the full rubric. Faculty Mentor will now explain exactly how your decision will be scored. No surprises during the game.

• If Faculty Mentor has not already transitioned to scoring, type: "Explain the scoring system. How does weakest-link logic work, what does the 0-100 scale mean, and what are the 12 governance markers?"

Faculty Mentor will cover three things:

• Weakest-link logic: Your overall decision quality equals your lowest single element score. Score 95 on five elements but 40 on one? Your decision quality is 40.
• The 0-100 scale: 0-49 means not addressed or fundamentally flawed. 50-69 means attempted but significant gaps remain. 70-84 means solid and meets minimum bar. 85-100 means excellent and board-ready.
• 12 governance markers: The Board Director persona will score your final decision memo against 12 specific criteria (problem statement, alternatives, recommendation, financial impact, risks, stakeholder impact, timeline, resources, success metrics, reversibility, competitive response, and ESG/reputation). You need at least 10 of 12 addressed to pass.

Why see the rubric before playing?

In Stanford's executive education, participants see the evaluation criteria upfront. This forces you to aim at the standard rather than guess at it. You now know exactly what 70/100 looks like on each element, and you know the 12 markers your memo must address.

• Confirm with Faculty Mentor that you understand the scoring by typing: "I understand the scoring. The passing bar is 70 on every element, weakest-link logic, and 10 of 12 governance markers. Ready to move on."

Session feeling too long or losing coherence?

If the Cowork session starts repeating itself or loses track of where you are in the Framework Crash Course, type: "Let's recap. Where are we in the Framework Crash Course phase? I've covered [list what you've done]. What's next?" This re-anchors Claude to the system prompt's phase structure.

help me re-anchor my Cowork session to the Framework Crash Course phase

You now have both frameworks loaded in memory. You know how you will be scored. Next up, Faculty Mentor will demo both frameworks on a real decision you actually care about, so you can see the method in action before the pressure starts.

Warm Up With Your Real Decision

You now know the frameworks intellectually. You can name all six DQ elements and recite Pfeffer's rules. But knowing a framework and using a framework under pressure are completely different skills.

Before you face the NorthernTech scenario (where you'll be scored and gated), Faculty Mentor will model what rigorous DQ + Power analysis looks like on a decision you actually care about. This is your safe practice run.

In this step, get ready to:

• Share a real pending decision with Faculty Mentor and watch it get deconstructed.
• See all 6 DQ elements applied to your decision with a partial scorecard demonstration.
• Experience the Power overlay on something personal so the NorthernTech scenario feels natural.

Bring your real decision to Faculty Mentor

The warm-up works best when you bring a decision you genuinely face right now. Not a hypothetical. Not something from last year. A real choice where you have not yet committed.

What counts as a real decision?

Think of something where you have two or more meaningfully different paths and genuine uncertainty about which to choose. Career moves, team restructures, product bets, investments, partnerships, hiring decisions, or organizational changes all work well.

The higher the stakes for you personally, the more visceral the DQ + Power analysis will feel. Faculty Mentor will identify gaps you have not noticed.

• In your Claude Desktop Cowork session, start a new task by typing the following prompt:

Run the Warm-Up Demo phase from the system prompt. Faculty Mentor walks me through both frameworks using a real decision I'm facing. Here's my decision:

[[YOUR_DECISION="Describe your real pending decision in 2-3 sentences. Name the options you're considering and why you're stuck."]]

Start by asking clarifying questions to frame this properly.

What does this prompt do?

This activates Phase 3 (Warm-Up Demo) from the system prompt you loaded in Step 1. Faculty Mentor will ask 2-3 clarifying questions before diving into analysis. Answer honestly. The more specific you are, the more useful the demo becomes.

• Answer Faculty Mentor's clarifying questions with real details about your situation.
• Do not sanitize or simplify. Include the messy constraints (budget, people, timeline, politics) that make the decision hard.

Keep it to 2-3 exchanges

Faculty Mentor should ask at most 2-3 clarifying questions. If it keeps asking, nudge it forward: "You have enough context. Walk me through the DQ elements now." This phase has a 13-minute budget.

Watch the DQ walkthrough on your decision

Once Faculty Mentor has enough context, it walks your decision through all 6 DQ elements out loud. This is not you doing the work yet. This is Faculty Mentor modeling what rigorous process looks like so you can replicate it in the live scenario.

• Read Faculty Mentor's analysis carefully as it works through each element.
• Pay attention to where Faculty Mentor says your framing is too narrow or where you are missing alternatives.
• Notice how it identifies information gaps you have not named. These are pieces of information that, if known, would change your decision.

What should the partial scorecard look like?

Faculty Mentor will score a few elements of your decision to demonstrate the 0-100 scale in action. Expect to see at least one element land below 70 (the minimum passing bar). This is intentional. It shows you what "gap identified but no plan to close it" looks like versus "key gaps named with a plan."

Faculty Mentor is not grading you here. It is showing you the difference between the scoring bands so you can self-assess during the NorthernTech exercise.

• If Faculty Mentor does not produce a partial scorecard after walking through all 6 elements, prompt it: "Score at least 3 of those elements 0-100 to show me how the scale works on my decision."

Faculty Mentor being too generic?

If the analysis feels like a template rather than specific to your situation, push back: "You're being generic. Name the specific alternatives I haven't considered for MY decision. What specific information gap would change MY choice?" The system prompt is designed to respond to direct challenges with sharper specificity.

help me get Faculty Mentor to give more specific, pointed DQ analysis of my decision instead of generic advice

Experience the Power overlay on your decision

After the DQ walkthrough, Faculty Mentor shifts to Pfeffer's lens. This is where the analysis gets uncomfortable. Power analysis names who wins, who loses, and what political moves each option requires.

• If Faculty Mentor does not transition to the Power overlay automatically, prompt it: "Now apply Pfeffer's Rules to my decision. Who benefits from each option? Who loses? What does each choice do to my brand?"
• Listen for Faculty Mentor's analysis on three specific rules:

• Rule 4 (Build a Powerful Brand): What does each option signal about you as a decision-maker?
• Rule 5 (Network Relentlessly): Which relationships do you need to cultivate or activate for each path?
• Rule 6 (Use Your Power): Is there a first-mover advantage where acting now matters more than waiting?

Why does this matter before NorthernTech?

Seeing both frameworks applied to something you care about makes the NorthernTech scenario feel like a method you are practicing, not a case study you are reading. When Roy pushes back on your NorthernTech decision in the next step, you will already know what Power-aware reasoning feels like.

• After the Power overlay, ask yourself: did the Power analysis change which option I would choose, or only how I would implement it?
• Hold that question in mind. Faculty Mentor will ask you the same thing about NorthernTech in Step 5.

You have now seen both frameworks applied to a decision that matters to you personally. The NorthernTech scenario is next. This time, you are not watching. You are the General Manager making a $400M decision under time pressure, with Roy pushing back and Board Director waiting to score you.

Run the Live NorthernTech Decision

Your warm-up round proved that DQ and Power aren't abstract theories. They changed how you see a decision you actually care about. Now the stakes go up.

You're about to step into a 30-minute live scenario as the General Manager of NorthernTech's $400M business unit. An AI-native competitor is eating your customers. Churn has nearly doubled. The board meets in three weeks. Faculty Mentor guides, Roy challenges, and you decide.

In this step, get ready to:

• Frame the NorthernTech problem and generate creative alternatives beyond the obvious three paths.
• Weigh information gaps and values while Roy surfaces real operational constraints.
• Score all 6 DQ elements on your completed scorecard.

Frame the problem and generate alternatives (DQ Elements 1-2)

The most common mistake in strategic decisions is treating a solution as the problem statement. "We need to add AI" is a solution disguised as a frame. "Our iteration speed is 72 hours slower than VectorFlow, causing 11% churn" is a real frame that opens different alternatives.

Faculty Mentor will push you on this distinction hard. If your frame is too narrow, your alternatives collapse. If it's too broad, you'll never decide.

• In Claude Desktop, start a new Cowork task by clicking the text input field at the bottom of the Cowork panel.
• Paste the following prompt to activate the Live Decision phase:

Run the Live Decision phase (Phase 4) from the system prompt in system-prompt.md.

Activate the full NorthernTech scenario. Faculty Mentor guides me through DQ Elements 1-6 in sequence. Roy enters at Element 3 (Information) with his operational constraints.

Start with Element 1: Appropriate Frame. Challenge my framing. Do not let me disguise a solution as a problem statement.

Write progressive updates to decision-memo.md as we work through each element. At the end, write my scores to dq-scorecard.md.

Enforce time pressure: remind me at the equivalent of 60 min, 75 min, and 85 min marks.

What happens when you send this prompt?

Claude reads your system-prompt.md file and activates the Faculty Mentor persona. It presents the full NorthernTech scenario brief and immediately asks you to articulate the problem frame.

The simulator treats this as a 30-minute real-time exercise. Faculty Mentor will push back, flag bias traps, and refuse to advance until your framing is crisp.

Faculty Mentor opens with the scenario brief and asks you to frame the decision. Here is what to expect and how to engage:

• Articulate the problem in business-impact terms. "How do we stop the churn bleeding" produces different alternatives than "how do we add AI." State who is affected and what is NOT in scope.
• When Faculty Mentor asks for alternatives, push beyond the three obvious paths (Build, Buy, Partner). Generate at least one sequenced option (Partner-then-Build), one partial approach (acqui-hire VectorFlow's CTO team only), or one unconventional path.
• If Faculty Mentor calls out anchoring bias (your alternatives cluster around your gut prediction), take the flag seriously. Name an alternative that directly contradicts your gut call.

Treat this like a real board prep

The more honestly you engage with the scenario, the stronger your learning signal. If you catch yourself giving "safe" answers that you think the simulator wants to hear, stop. Say what you actually believe. Faculty Mentor is designed to challenge, not to validate.

Faculty Mentor not pushing back on your frame?

If the responses feel generic or overly agreeable, reply with: "You're being too easy on me. The system prompt says you refuse MBA jargon and flag bias traps. Challenge my framing harder." This re-anchors the persona.

help me get Faculty Mentor to challenge my decision framing more rigorously

Weigh information gaps and values as Roy enters (DQ Elements 3-4)

Most people gather information that confirms their preferred option. Faculty Mentor flips this: "What information would CHANGE your decision? Not what would confirm it." This is the hardest question in Decision Quality because it requires imagining a world where you're wrong.

This is also where Roy enters the exercise. Roy is a Director-level leader with 40 engineers. He brings information you don't have and constraints you haven't considered.

• When Faculty Mentor transitions to Element 3 (Information), name at least 3 specific information gaps. For each gap, state whether knowing the answer would change your decision or only confirm it.
• Roy will surface 4 critical constraints you didn't know about. Listen carefully. These change the math on every path.
• When Faculty Mentor transitions to Element 4 (Values), explicitly state your trade-offs using the format: "I am choosing X OVER Y because Z." Name at least 3 competing values (speed vs. control vs. talent retention vs. cost vs. risk).

What does Roy reveal?

Roy brings four pieces of information that aren't in the scenario brief. Six of his best engineers specifically joined NorthernTech to build AI capabilities. If you choose Buy or Partner, they may leave.

VectorFlow's CTO is a former NorthernTech engineer who left over "lack of innovation vision." The Partner path requires VectorFlow to deprioritize 3 other potential partners, meaning they'll want equity or a board seat. The Build path assumes you can hire 25 AI engineers in a market where top talent gets 4-5 competing offers.

What are bias traps?

Faculty Mentor is programmed to fire specific interventions when it detects cognitive biases. If you say "we've already invested X in our platform," expect a sunk cost challenge. If you only gather confirming information, expect a confirmation bias flag. These interruptions are the exercise working, not a sign you're doing poorly.

• After Roy delivers his constraints, respond with how those constraints change your evaluation. Do not dismiss them. Integrate them into your values weighting.

Complete reasoning and commitment, then score your decision (DQ Elements 5-6)

Sound Reasoning is where most intuitive decision-makers get caught. You may feel certain about your choice, but can you connect premises to conclusion in a way that survives adversarial challenge? Faculty Mentor checks for logic gaps and intuitive leaps disguised as analysis.

Commitment to Action is where the decision becomes real. A decision without named stakeholders, Day 1 actions, and an execution plan is a wish, not a choice.

• When Faculty Mentor transitions to Element 5 (Reasoning), state your key assumptions explicitly. For each one, answer: "If this assumption is wrong, does my conclusion change?"
• Run a pre-mortem when prompted: "If this decision fails in 18 months, the most likely cause is ___." Name at least 2 failure modes.
• For Element 6 (Commitment to Action), name the specific stakeholders who must commit. Describe what Day 1 looks like. Give Roy enough detail that he can brief his 40 engineers on Monday morning.

What does Roy's stress-test sound like?

Roy won't let you off easy on Commitment. If you choose Buy, expect: "Give me the 60-day integration plan right now or your Commitment to Action score is below 70." If you choose Build, expect: "Where are 25 AI engineers coming from? Name the recruiting timeline." If you choose Partner, expect: "What does my team DO for 3 years while VectorFlow builds the AI layer?"

• After completing all 6 elements, Faculty Mentor asks you to score each element 0-100. Be honest. Remember the scoring rubric: below 50 means not addressed, 50-69 means attempted but gaps remain, 70-84 is solid, and 85-100 is excellent.
• Cowork writes your scores and rationale to dq-scorecard.md in your project folder. It also updates decision-memo.md with the DQ Six Elements Summary section filled in.

What is weakest-link scoring?

Your overall decision quality equals your lowest single element score. A decision that scores 95 on five elements but 40 on one element is a 40-quality decision. The chain breaks at its weakest link. This is why Faculty Mentor pushes hard on every element instead of letting you coast through the ones that feel easy.

• After Cowork writes the files, verify that dq-scorecard.md exists in your stanford-sep-simulator folder with all 6 elements scored.
• Open dq-scorecard.md in your text editor. Confirm each element has a numeric score and a rationale. The weakest link section should identify your lowest-scoring element.

Scorecard file not appearing in your folder?

If Cowork didn't write the file automatically, type this in the chat: "Please write my DQ scorecard scores to dq-scorecard.md now, including rationale for each element and the weakest link analysis." Cowork may have been waiting for explicit confirmation before writing.

If the file exists but is empty or incomplete, check that Claude Desktop still has write access to your folder. Click the folder icon in Cowork to re-confirm permissions.

help me get Cowork to write my DQ scorecard file

✔️ I see my scorecard with all 6 scores

Your scorecard is locked. Note your weakest-link score. If any element is below 70, the Standards Gate in Step 6 will catch it and you'll rework that element. For now, move forward.

ⓧ Cowork lost context mid-exercise

Long Cowork sessions can lose context. If the responses became generic or the simulator stopped enforcing phases, paste this re-anchoring prompt:

Re-read system-prompt.md in this folder. We are in Phase 4 (Live Decision), Element [NUMBER] of the DQ Six Elements. Resume the Faculty Mentor persona with full anti-drift guardrails. Roy should be active if we are on Elements 3-6.

Here is where I am: [briefly state your current position in the exercise]

Replace [NUMBER] with the element you were working on (1-6) and fill in your current position. Claude re-reads the system prompt from disk and resumes.

You now have a scored decision backed by rigorous process. But DQ alone misses something critical: power dynamics. Next, Faculty Mentor applies Pfeffer's 7 Rules to test whether your decision survives implementation.

Apply the Power Overlay

You now have a DQ-backed decision with all six elements scored. That analysis tells you the rigorous answer. But rigorous answers die in organizations every day because they ignore a critical question: who has the power to kill this, and who has the power to champion it?

In this step, Faculty Mentor switches from DQ mode to Pfeffer's Power lens. You will examine your chosen path through the political reality of organizational dynamics. The result is a power-overlay.md file that maps who benefits, who loses, and what implementation moves your decision requires to survive contact with real people.

In this step, get ready to:

• Apply Pfeffer's 7 Rules of Power to your NorthernTech decision.
• Determine whether power dynamics change your decision or only your implementation approach.

Activate Pfeffer's lens on your decision

DQ measures whether your reasoning is sound. Power measures whether your decision will survive implementation. A decision that ignores who benefits and who loses inside the organization is a decision that dies in a meeting room.

• In your active Cowork session, start a new task with this prompt:

Run the Power Overlay phase from the system prompt. Faculty Mentor switches to Pfeffer's lens now.

For my chosen path, I need to answer:
- Who benefits from this choice inside the organization?
- Who loses, and how are the losers addressed?
- What does this decision do to MY brand (Rule 4)?
- Does this decision require me to Network (Rule 5) with people I haven't cultivated yet?
- Is there a first-mover advantage where using power now (Rule 6) matters?

Write the full power-overlay analysis into power-overlay.md when we finish.

What happens during this phase?

Faculty Mentor will ask you pointed questions about the political landscape of your decision. These are not abstract. For the NorthernTech scenario, think about Roy's 40 engineers, the CEO you report to, the board that meets in 3 weeks, and VectorFlow's CTO (a former NorthernTech engineer).

Your job is to answer honestly for your chosen path. If you chose Buy, who inside NorthernTech benefits from that acquisition? If you chose Build, what does that signal about your brand as a leader? Faculty Mentor will push back if your answers are shallow.

• Answer Faculty Mentor's questions about each Power dimension for your chosen path.
• Be specific. Name actual stakeholders from the scenario (Roy, the CEO, VectorFlow's CTO, the board) rather than abstract roles.
• When Faculty Mentor completes the analysis, confirm that power-overlay.md has been written to your stanford-sep-simulator folder.

Open power-overlay.md in your text editor. You should see sections covering Who Benefits, Who Loses, and entries for Rules 1, 4, 5, and 6 applied to your specific decision.

Why only some of the 7 Rules?

The Power Overlay focuses on Rules 1, 4, 5, and 6 because these are the rules most directly relevant to a strategic decision's survival. Rule 1 (Get Out of Your Own Way) catches self-sabotage. Rule 4 (Brand) checks what the decision signals. Rule 5 (Network) identifies who you need aligned. Rule 6 (Use Your Power) tests for first-mover advantage.

Rules 2, 3, and 7 apply more to execution style than decision selection. They will come into play when you brief stakeholders and implement.

Faculty Mentor not switching to Power mode?

If the conversation is still focused on DQ elements or the scenario details, explicitly say: "I'm ready for the Power Overlay phase. Please switch to Pfeffer's lens." If Cowork has lost context after the long Step 4 session, reference the system prompt directly: "Read the Power Overlay section in system-prompt.md and run that phase now."

Still stuck? Why isn't Faculty Mentor switching to the Power Overlay phase?

Run the integration check

This is the critical junction. Faculty Mentor will ask you directly: did the Power overlay change your decision or only your implementation approach? The answer determines what happens next.

• When Faculty Mentor asks the integration question, answer honestly based on what the Power analysis revealed.

What counts as a decision change vs. implementation change?

If discovering that Roy's best engineers will leave (a power/stakeholder dynamic) makes you switch from Buy to Partner, that is a decision change. You must re-score your DQ elements because the alternative you chose has changed.

If your decision stays the same but you now realize you need to have a private conversation with Roy before the board meeting (a network move), that is an implementation change. Your DQ scores stay the same. You add implementation notes to the memo.

• If Power changed your decision, tell Faculty Mentor your new chosen path and ask Cowork to update dq-scorecard.md with revised scores and a revision history entry.
• If Power only changed your implementation approach, tell Faculty Mentor what adjustments you will make and confirm implementation notes are added.
• Ask Cowork to update the Pfeffer Power Overlay section of decision-memo.md with the full power analysis by prompting:

Update decision-memo.md with the Pfeffer Power Overlay section. Fill in all six fields:
- Who benefits
- Who loses and how addressed
- Brand implications (Rule 4)
- Network moves required (Rule 5)
- First-mover considerations (Rule 6)
- Implementation approach adjusted by Power lens

Also update the Integration Check section in power-overlay.md with whether Power changed my decision or only implementation.

What does this prompt do?

This prompt ensures both output files are complete. power-overlay.md captures the full analysis. The Pfeffer Power Overlay section in decision-memo.md captures the summary that the Board Director will review in the next step.

• Verify both files exist and are populated by opening power-overlay.md and decision-memo.md in your text editor.
• In decision-memo.md, confirm the Pfeffer Power Overlay section has all six bullet points filled with specific content (not blanks or placeholders).
• In power-overlay.md, confirm the Integration Check section states whether Power changed your decision or only implementation.

Files not updating?

If Cowork confirms writing but the files appear empty or unchanged in your text editor, close and reopen the file. Some text editors cache the old version. If the files genuinely were not written, ask Cowork directly: "Please write the power-overlay.md and update decision-memo.md now. Show me the file contents you're writing."

Need more help? Cowork isn't writing my power-overlay.md file

Your decision now has both analytical rigor (DQ) and political awareness (Power). Next up, the Board Director enters to score your complete memo against 12 governance markers. This is where you find out if your work is board-ready or needs rework.

Pass the Standards Gate Review

You have a DQ-scored decision with a Power overlay applied. In a real Stanford Executive Program cohort, this is where your peers and faculty would pressure-test your memo before it reaches the board.

In this step, the Board Director persona enters as a governance gatekeeper. The Director scores your memo against 12 specific markers, then Roy checks whether your decision is concrete enough to brief his engineering team on Monday morning.

In this step, get ready to:

• Have Board Director score your memo against 12 governance markers.
• Rework any gated elements until all DQ scores hit 70+ and at least 10 markers are addressed.
• Get Roy's final adoption verdict confirming your memo is briefable.

Activate the Board Director

The Board Director persona scores without encouragement or coaching. It reads your decision-memo.md and dq-scorecard.md, then grades each of the 12 governance markers as Addressed, Not Addressed, or Partially Addressed.

• In Claude Desktop, start a new Cowork task with this instruction:

Run the Standards Gate Review phase from the system prompt.

Board Director: Read my decision-memo.md and dq-scorecard.md.
Score each of the 12 governance markers criterion-by-criterion.
Then check if any DQ element in my scorecard is below 70/100.

Give me the full scoring before any commentary.

What does Board Director look for?

The 12 governance markers test whether your memo is board-ready. They cover problem statement clarity, financial impact, risk assessment, stakeholder mapping, timeline, resource requirements, success metrics, reversibility, competitive response, and ESG exposure.

Any DQ element below 70/100 triggers a gate. Any governance marker scored "Not Addressed" gets flagged. Most first attempts get gated on at least one element.

Board Director will present scores without embellishment. You will see each marker listed with its status, plus any DQ elements that fall below the 70/100 threshold.

Expect to get gated

Getting gated is not failure. It is the exercise working. The rework cycle is where the deepest learning happens because it forces you to articulate what was implicit in your reasoning.

Rework any gated elements

If Board Director gates you, Faculty Mentor re-enters to coach you through your weakest element. This is a dialogue, not a lecture. You rework the element, then Board Director re-scores.

• Read Board Director's scoring carefully. Identify which DQ elements scored below 70 and which governance markers are Not Addressed.
• Tell Cowork to proceed with rework by typing this instruction:

Faculty Mentor: Coach me through my weakest element.
Help me strengthen it to 70+ without doing the work for me.

After I rework it, have Board Director re-score.
Update dq-scorecard.md with the revision and update decision-memo.md with any new content.

Continue the rework cycle until all 6 elements are at 70+ and at least 10 of 12 markers are Addressed.

How does the rework cycle work?

Faculty Mentor asks you targeted questions about the weak element. You answer with specifics. Faculty Mentor may point out where your reasoning has gaps or where you are being vague. Once you have strengthened your response, Board Director re-scores that element.

The cycle repeats until every DQ element hits 70/100 minimum and at least 10 of the 12 governance markers show as Addressed. This is the weakest-link rule in action.

• Respond to Faculty Mentor's coaching questions with concrete answers. Name specific numbers, timelines, stakeholders, and actions.
• After each rework, confirm Board Director's new score before moving to the next weak element.
• Continue until Cowork confirms all elements are at 70+ and at least 10 markers are Addressed.

Board Director keeps gating you?

The most common weak spots are Commitment to Action (missing Day 1 specifics or stakeholder names) and Meaningful Information (listing data you have instead of naming what would change your decision). If you are stuck, ask Faculty Mentor: "What specific sentence would raise this element above 70?"

help me understand why my element keeps scoring below 70 and what specifically I need to add

Get Roy's adoption verdict

Once Board Director passes your memo, Roy gives the final operational test. Roy's question is simple: "Can I brief my team Monday morning based on this memo?" If the answer is no, Roy names exactly what is missing.

• After Board Director confirms gate status as PASS, tell Cowork to activate Roy's adoption verdict:

Roy: You've seen my decision memo. Can you brief your team of 40 engineers Monday morning based on this?

If no, tell me exactly what's missing. I'll add it.
Once you say YES, update decision-memo.md with your verdict in the Roy's Adoption Verdict section.

What does Roy usually flag?

Roy typically pushes back on three things: timeline specificity ("When does my team start and what do week 1 actions look like?"), talent plan ("Which engineers move where and who backfills?"), and first-week actions ("What do I tell people on Monday morning?"). These are execution details that strategists often skip.

• If Roy says no, add the missing specifics he names. Then ask Roy to re-evaluate.
• Continue until Roy confirms YES. Cowork writes the verdict into decision-memo.md.

• Verify that Cowork has updated both files by checking your stanford-sep-simulator folder.
• Open decision-memo.md and confirm the 12 Governance Markers Checklist, Roy's Adoption Verdict (YES), and Board Director Score (PASS) sections are all populated.
• Open dq-scorecard.md and confirm the Revision History section includes a "Post-Standards-Gate revision" entry with all elements at 70+.

Files not updated?

If Cowork did not write the results to your files, tell it explicitly: "Write the Board Director scores and Roy's verdict into decision-memo.md. Update dq-scorecard.md revision history with the post-Standards-Gate scores." Cowork sometimes needs a direct file-write instruction after a long interactive exchange.

help me get Cowork to write the Standards Gate results into my files

Your decision memo has now survived a governance gate and an operational adoption test. In the next step, you will debrief the entire exercise, record a teach-back proving you can explain both frameworks from memory, and seed your first real-world decision for the next 7 days.

Debrief, Teach Back, and Commit

Your decision memo passed the Standards Gate. Board Director signed off, Roy can brief his team Monday, and every DQ element scored 70 or above. The decision is shipped.

Now comes the part that separates a single exercise from a lifetime skill. You will extract the learning signal from your predict-then-decide delta, prove you own both frameworks by teaching them from memory, and plant a transfer seed that forces you to apply DQ + Power to a real decision within 7 days.

In this step, get ready to:

• Run an After-Action Review comparing your gut prediction to your DQ-backed decision.
• Record a teach-back video proving you can explain both frameworks from memory.
• Build your Personal Decision Quality Operating System with a real decision calendared for the next 7 days.

Run the After-Action Review

The predict-then-decide pattern creates a measurable learning signal. The gap between your gut call (locked before any framework knowledge) and your final DQ-backed decision reveals exactly which elements of the methodology changed your thinking.

Why does the delta matter?

If your final decision matches your gut prediction, one of two things is true: you have exceptional intuition for this type of decision, or the frameworks did not challenge you hard enough. Either way, naming which DQ element or Power rule most shifted your thinking tells you where the methodology adds value for YOU specifically.

Faculty Mentor will walk you through a structured debrief. This is not a victory lap. It is a rigorous comparison of before vs. after.

• In Claude Desktop, start a new Cowork task with the following instruction:

Run the After-Action Review phase from the system prompt.

Faculty Mentor debriefs me:
1. Compare my gut prediction to my DQ-backed decision - what happened?
2. Which DQ element most changed my thinking?
3. Which Power rule most changed my implementation approach?
4. What would I do differently if I had 2 hours instead of 30 minutes?

Write the full AAR into a file called aar-notes.md in this folder.
Also update the Predict-Then-Decide Delta section at the bottom of decision-memo.md.

What happens during the AAR?

Faculty Mentor reads your locked gut prediction from decision-memo.md and compares it to your final recommendation. The debrief focuses on three questions: what changed, why it changed, and what you would improve with more time.

Faculty Mentor also names any bias traps that fired during the exercise (anchoring, sunk cost, confirmation bias, status quo bias, or authority bias). This self-awareness is part of the learning signal.

• Engage honestly with Faculty Mentor's questions. Answer from memory before looking at your files.
• When Faculty Mentor asks what you would do differently, think about which DQ element felt rushed or which information gap you wish you had closed.

• Verify that aar-notes.md now exists in your stanford-sep-simulator folder.
• Open decision-memo.md and scroll to the bottom. Confirm the Predict-Then-Decide Delta section is now filled in with your gut prediction, final decision, what changed, and key learning.

Don't see aar-notes.md in your folder?

Make sure Claude Desktop is still connected to your stanford-sep-simulator folder in Cowork mode. If the session disconnected, click Work in a Folder again and re-select the folder. Then re-run the AAR task.

help me troubleshoot why Cowork is not writing files to my folder

Record your teach-back

Teaching from memory is the strongest test of internalization. If you can explain both frameworks to an imaginary colleague without looking at your reference cards, you own the methodology.

Why record instead of just summarize in text?

Speaking out loud forces you to organize your thoughts sequentially. You cannot hide behind bullet points or copy-paste. The recording also becomes a future reference you can revisit before your next real decision.

• Open Loom (free tier, 5-minute limit per video) in your browser, or open your phone camera app.
• Close or minimize all your reference files. You are explaining from memory only.
• Start recording and explain the following four things to an imaginary colleague:

• What are the DQ Six Elements? Name all six.
• How does weakest-link scoring work? Explain why a decision with five 95s and one 40 is a 40-quality decision.
• What do Pfeffer's Rules add that DQ alone misses? Explain the integration logic (rigor first, power second).
• What decision did you make for NorthernTech and why? Summarize your recommendation in 30 seconds.

• Stop recording when finished. Aim for 3-5 minutes total.

Use the DQ hexagon as a mental anchor

Visualize the six elements as points on a hexagon. Start at Frame (top), move clockwise through Alternatives, Information, Values, Reasoning, Commitment. This spatial memory trick makes recall much easier under pressure.

Now bring your summary back to Faculty Mentor for review.

• In Claude Desktop, type a brief summary of what you covered in your recording. Tell Faculty Mentor which elements you named and how you described the integration logic.

Faculty Mentor: I just recorded my teach-back. Here's what I covered from memory:

- The six DQ elements I named: [list them]
- How I explained weakest-link scoring: [brief summary]
- How I explained what Power adds to DQ: [brief summary]
- My NorthernTech recommendation summary: [brief summary]

Please review and flag anything I misstated or omitted.

What does Faculty Mentor check?

Faculty Mentor compares your summary against the canonical definitions. If you misordered the elements, confused one rule with another, or omitted the integration logic (DQ rigor first, Power second, never in isolation), Faculty Mentor flags it so you can re-record or make a mental correction.

• If Faculty Mentor flags any gaps, review the specific point and consider re-recording that section.
• Save your Loom recording or transfer the phone video to your computer for future reference.

Build your Transfer Seed

Frameworks without application deadlines decay within 48 hours. Research on training transfer shows that naming a specific future decision and setting a calendar trigger dramatically increases the odds you will actually use what you just learned.

You will now build your Personal Decision Quality Operating System. This is a 1-page reusable framework that captures the full 10-phase method plus your first real-world decision queue.

• Think of ONE specific decision you face in the next 7 days. This must be a real decision, not a hypothetical. Career move, team restructure, product bet, investment, partnership, hiring call. Something with genuine stakes.
• Tell Claude your decision and instruct Cowork to finalize your Operating System file:

Run the Transfer Seed phase from the system prompt.

My real decision for the next 7 days is: [describe your decision in one sentence]

Please:
1. Create operating-system.md (based on the template) with the full 10-phase method
2. Fill in my 7-Day Decision with the decision I just named and the date (7 days from today)
3. Fill in the 30-Day Re-Score with a date 30 days from today
4. Add NorthernTech as Decision 1 in the Running Log (using my gut prediction, final decision, weakest element, and key learning from the AAR)
5. Add my named real decision as Decision 2 in the Running Log (fields blank, to be filled in 7 days)
6. Mark both calendar events as YES

What goes into the Operating System?

The operating-system.md file captures the full 10-phase DQ + Power method in a condensed, actionable format. It also includes your decision queue (7-day and 30-day triggers) and a running log where you track every decision you run through the method over time.

This is the file you open every time you face a significant decision going forward. The method is invariant. Your decisions change. The rigor stays.

• Verify that operating-system.md exists in your folder (not operating-system-template.md, which is the blank template).
• Open operating-system.md and confirm your named decision appears in the 7-Day Decision section with a date 7 days from today.
• Confirm the Running Log shows NorthernTech as Decision 1 (filled) and your real decision as Decision 2 (blank, awaiting future use).

Now set the actual calendar events that will trigger your future application.

• Open your calendar app (Google Calendar, Outlook, Apple Calendar, or any app you check daily).
• Create an event 7 days from today with the title: Run DQ + Power on [your decision name].
• Create a second event 30 days from today with the title: Re-score: did I use DQ + Power on at least one real decision this month?.
• In the 30-day event description, add: Which element was weakest? Open operating-system.md and fill in the Running Log..

Set a reminder notification

Add a 1-hour-before reminder to both events. The 7-day event should prompt you to block 40 minutes for the full DQ + Power method on your real decision. You can re-use the simulator by starting a fresh Cowork task referencing your system-prompt.md file.

Not sure which decision to pick?

Choose a decision where at least two people would disagree on the right path. If everyone would agree, it is not a real decision. Good candidates: whether to pursue a new role, how to allocate your team's next sprint, whether to invest in a tool or build internally, or which candidate to hire.

help me identify a good real decision for my 7-day transfer commitment

• Take a final look at your stanford-sep-simulator folder. You should now have 8 files total:

• system-prompt.md (the simulator engine).
• dq-six-elements-reference.md (framework reference card).
• pfeffer-seven-rules-reference.md (framework reference card).
• decision-memo.md (your completed NorthernTech board memo with Predict-Then-Decide Delta filled in).
• dq-scorecard.md (your finalized scores, all 70+).
• power-overlay.md (Pfeffer's lens applied to your decision).
• aar-notes.md (your After-Action Review debrief).
• operating-system.md (your lifetime-reusable DQ + Power method with decision queue).

Secret mission

Run a Second Scenario Through the Full Simulator

You proved the method works on NorthernTech. But can you do it again on a completely different strategic decision in half the time? Choose a second scenario and run the full DQ + Power pipeline to prove this is now a repeatable skill, not a one-time exercise.

Clean Up Your Resources

Clean Up Your Resources

This project runs entirely locally using your existing Claude Pro subscription, so there are no ongoing costs beyond what you already pay. Decide whether to keep your resources as-is, pause them to come back later, or delete them entirely.

Resources you used:

• Local folder stanford-sep-simulator with 6+ markdown files (system prompt, reference cards, templates, output files).
• Claude Desktop App session (Cowork mode) with conversation history.
• Loom free account (if used for teach-back recording).
• Calendar events (7-day and 30-day transfer commitments).

Keep everything running

No action needed. Choose this if you plan to run additional scenarios through the simulator or want to refine your decision memo further.

• Your stanford-sep-simulator folder stays on your Desktop with all files intact.
• Your Cowork session history remains accessible in Claude Desktop.
• Your calendar events stay scheduled for the 7-day and 30-day check-ins.
• You can reopen Cowork, point it at the same folder, and run new scenarios anytime.

Pause - I'll come back to this later

Keep your files but close active sessions to free up system resources.

• Close your Cowork session in Claude Desktop. Your conversation history is saved automatically.
• Keep the stanford-sep-simulator folder where it is. All your output files (decision memo, scorecard, AAR notes) remain accessible.
• Keep your calendar events active so you still get the 7-day and 30-day reminders.
• When you return, open Claude Desktop, switch to Cowork, and point it at your existing folder to resume.

Delete - I don't want to use this again

Remove all project files and accounts you no longer need.

• Delete the stanford-sep-simulator folder from your Desktop.
• In Claude Desktop, close and delete the Cowork session history if you want a clean slate.
• Delete your Loom account if you created one only for this project (go to Loom settings, then account deletion).
• Remove the 7-day and 30-day calendar events if you do not plan to continue practicing.
• Your Claude Pro subscription is unaffected. No changes needed there.

That's a Wrap!

That's a Wrap!

Nice work! You just practiced the core strategic decision-making methodology from Stanford GSB's Executive Program in a single 90-minute tabletop session.

You've learned how to:

• Build a reusable 3-persona decision simulator (Faculty Mentor, Roy, Board Director) that you can apply to any strategic decision going forward.
• Apply the DQ Six Elements with weakest-link scoring and Pfeffer's 7 Rules of Power to produce a board-ready decision memo passing 12 governance markers.
• Create your Personal Decision Quality Operating System with a named real decision queued and calendared at 7 and 30 days for transfer commitment.
• Secret Mission: Run a second scenario through the full simulator, proving the method generalizes beyond a single case.

Your predict-then-decide delta is your learning signal. If the DQ process did not change at least one thing about your initial gut call, either you have exceptional intuition or you did not let the framework challenge you hard enough.

Ready to quiz yourself?