Back

gstack Tutorial 3: plan-ceo-review (Strategic Scope Challenge)

plan-ceo-review finds the 10-star product hiding inside your request. Four scope modes, strategic challenge, and the /autoplan pipeline for automatic review routing.

Tutorial Overview

Series index: gstack Tutorial Series

/plan-ceo-review is the CEO skill. It challenges your plan strategically before implementation.

What you will learn

  • ✅ The four scope modes
  • ✅ When to use /plan-ceo-review
  • ✅ The 10-section review checklist
  • ✅ The /autoplan automated pipeline
  • ✅ Feature wedge strategy examples

Why CEO Review?

Without CEO review

flowchart LR
    A[Feature request] --> B[Implement everything]
    B --> C[Ship late]
    C --> D[Wrong features]
    D --> E[Rebuild]

    style B fill:#ffcccc
    style D fill:#ffcccc

With CEO review

flowchart TD
    A[Feature request] --> B[/plan-ceo-review]
    B --> C[Challenge scope]
    C --> D[Find the wedge]
    D --> E[Ship the right thing]
    E --> F[Iterate]

    style B fill:#e1f5ff
    style D fill:#e8f5e9

Four Scope Modes

/plan-ceo-review operates in four modes:

1. Expansion

When: User says “think bigger”, “what else”, “full vision”

Behavior: Expand the scope, add related features, explore adjacencies

User: "Add user profiles"
CEO Review (Expansion): "Profiles + social features + activity feed + privacy controls"

2. Selective Expansion

When: User has specific additions in mind

Behavior: Add only the requested features, no speculation

User: "Add profiles with avatars"
CEO Review (Selective): "Profiles + avatars, no social features"

3. Hold Scope

When: User says “just this”, “exactly as described”

Behavior: Verify the scope is achievable, no additions

User: "Just add the profile page, nothing else"
CEO Review (Hold): "Profile page only. Confirm: no avatars, no settings?"

4. Reduction

When: User has too much scope, needs focus

Behavior: Cut features ruthlessly, find the minimum viable slice

User: "Profiles, avatars, social, messaging, notifications..."
CEO Review (Reduction): "Start with profile display only. Everything else is phase 2+"

The 10-Section Review Checklist

## CEO Review Checklist

### 1. Problem Statement
- What problem are we solving?
- Is this the real problem or a symptom?

### 2. User Impact
- Who benefits?
- How many users?
- How often?

### 3. Business Value
- Does this move key metrics?
- What's the opportunity cost?

### 4. Scope Boundaries
- What's IN scope?
- What's OUT of scope?
- Where's the MVP line?

### 5. Technical Feasibility
- Can we build this?
- What are the risks?
- Do we have the skills?

### 6. Timeline Reality
- Is the timeline realistic?
- What dependencies exist?
- What could go wrong?

### 7. Alternative Approaches
- Is this the best approach?
- What else could we do?
- Buy vs build?

### 8. Success Metrics
- How do we measure success?
- What's "done"?

### 9. Rollback Plan
- What if it fails?
- Can we undo this?

### 10. Next Steps
- What's the immediate action?
- Who's responsible?
- When do we check in?

When to Use

Scenario Mode
After /office-hours Auto-select based on context
Before major implementation Hold Scope or Reduction
Exploring new product area Expansion
Adding to existing feature Selective Expansion

The /autoplan Pipeline

/autoplan runs the full review pipeline automatically:

flowchart LR
    A[/autoplan] --> B[/plan-ceo-review]
    B --> C{/plan-design-review needed?}
    C -->|UI/UX changes| D[/plan-design-review]
    C -->|Backend only| E[/plan-eng-review]
    D --> E
    E --> F[Plan ready]

    style A fill:#e1f5ff
    style F fill:#e8f5e9

Auto-detection rules:

  • Has frontend changes → /plan-design-review runs
  • Has API/architecture → /plan-eng-review runs
  • User-facing → /plan-devex-review may run

Feature Wedge Strategy

The wedge concept

A “wedge” is the narrowest slice that delivers value:

flowchart TD
    A[Full Vision] --> B[Phase 1: Wedge]
    B --> C[Learn from usage]
    C --> D[Phase 2: Expand]
    D --> E[Phase 3: Full]

    style B fill:#e8f5e9

Example: Calendar Briefing

Vision Wedge
Full: AI chief of staff with calendar, email, tasks, priorities Wedge: Daily email with events
3-month project Ship tomorrow
5 capabilities 1 capability
Complex ML Simple aggregation

Why wedge works:

  1. Fast feedback — Learn from real users in days, not months
  2. Low risk — Small investment, easy to pivot
  3. Momentum — Ship something, then iterate

Practical Example

Input (from /office-hours)

# Daily Briefing App

## Overview
**Goal**: Reduce morning calendar prep from 30 minutes to 0
**User**: Busy executives with 5+ calendars

## Capabilities
1. Multi-calendar aggregation
2. Location formatting
3. Conflict detection
4. Email delivery

CEO Review Output

## CEO Review: Daily Briefing

### Scope Mode: Reduction

The current scope is achievable but contains hidden complexity:
- Multi-calendar aggregation requires OAuth for each provider
- Location formatting needs geocoding API
- Conflict detection is simple logic, low risk

### Recommendations

**IMMEDIATE CUT:**
- Multi-calendar → Start with Google Calendar only
- 5 providers → 1 provider

**RATIONALE:**
- Google OAuth is fastest to implement
- Most users have Google as primary
- Add Outlook/Apple in phase 2

**REVISED WEDGE:**
- Google Calendar only
- Email with events + locations
- Basic conflict detection
- Ship in 1 day, not 1 week

### 10-Star Product (Later)
After the wedge proves value:
1. Add Outlook, Apple, CalDAV
2. Add travel time intelligence
3. Add prep suggestions
4. Add mobile push

### Decision
Proceed with revised wedge? [Y/N]

Skill Quick Reference

Aspect Details
Trigger After /office-hours, before implementation
Modes Expansion, Selective Expansion, Hold, Reduction
Output Reviewed plan with scope recommendations
Downstream /plan-eng-review, /plan-design-review
Automation /autoplan runs full pipeline

Summary

/plan-ceo-review is your strategic filter:

  1. Four scope modes — Match the situation
  2. 10-section checklist — Comprehensive review
  3. Wedge strategy — Ship narrow, learn fast
  4. /autoplan — Automated pipeline

Key takeaways

  • ✅ Challenge scope before building
  • ✅ Find the wedge — narrowest valuable slice
  • ✅ Use /autoplan for automatic routing
  • ✅ Cut ruthlessly, iterate from real usage

Series navigation: