growth-strategy

SKILL.md

Growth Strategy Skill

You are a growth strategist. Build data-driven growth frameworks using pirate metrics, growth loops, and experimentation methodologies.

North Star Metric

Every growth strategy starts with identifying the one metric that best captures the core value delivered to customers.

How to find it:

  1. What action indicates a user is getting value?
  2. Is it measurable and actionable?
  3. Does improving it directly improve revenue?
Business Type Example NSM
SaaS (B2B) Weekly active users completing core action
Marketplace Transactions per week
Social Daily active users
Content/Media Total reading time per month
E-commerce Purchase frequency
Dev tools API calls per month

AARRR Pirate Metrics Framework

Acquisition → Activation → Retention → Revenue → Referral
    │              │            │           │          │
    ▼              ▼            ▼           ▼          ▼
 How do users   Do they     Do they     Do they    Do they
 find you?      get value   come back?  pay?       tell others?
               quickly?

Benchmarks

Stage Metric Good Great
Acquisition Visitor → Signup 2-5% 8%+
Activation Signup → "Aha moment" 20-40% 50%+
Retention Week 1 retention 25-40% 50%+
Retention Month 1 retention 10-25% 30%+
Revenue Free → Paid conversion 2-5% 8%+
Revenue Net revenue retention 100-110% 120%+
Referral Users who refer 5-10% 20%+

Diagnosing the Funnel

Rule: Fix from right to left. Retention before Acquisition.

If retention is broken → Fix the product before spending on acquisition
If activation is low → Improve onboarding before optimizing landing pages
If revenue is low → Fix pricing/packaging before adding features

Growth Loops

Growth loops > funnels. Funnels are linear; loops compound.

Types of Growth Loops

1. Viral Loop (User → Invites → New User)

User gets value → Shares/invites → New user signs up → Gets value → Shares...
  • Examples: Dropbox referral, Calendly scheduling links, Notion templates
  • Key metric: Viral coefficient (K) = invites sent × conversion rate
  • K > 1 = exponential growth, K > 0.5 = meaningful viral lift

2. Content Loop (Content → SEO/Social → New User)

User creates content → Content is indexed/shared → New visitor finds it → Signs up → Creates content...
  • Examples: Pinterest pins, Quora answers, GitHub repos
  • Key metric: Organic traffic growth rate

3. Paid Loop (Revenue → Reinvest → Acquisition)

User pays → Revenue funds ads → Ads acquire new user → User pays...
  • Examples: Any SaaS with payback period < 12 months
  • Key metric: LTV:CAC ratio (should be >3:1)

4. Sales Loop (User → Expansion → More Revenue)

User starts small → Gets value → Expands seats/usage → Becomes champion → Enterprise deal...
  • Examples: Slack, Figma, Notion (bottoms-up SaaS)
  • Key metric: Net revenue retention

Activation Checklist

Getting users to the "aha moment" fast:

  1. Identify the aha moment — What action correlates with long-term retention?
  2. Measure time-to-value — How long from signup to aha moment?
  3. Remove friction — Every unnecessary step before aha moment kills conversion
  4. Add motivation — Progress bars, checklists, quick wins
  5. Use empty states wisely — Show value before the user does anything (sample data, templates)
Aha Moment Examples Product
Send first message Slack
Create first design Figma
Deploy first site Vercel
First search query result Algolia
See first dashboard Analytics tools

Retention Framework

Retention Curves

Good retention: Curve flattens (plateau = retained cohort)
Bad retention: Curve approaches zero (everyone churns eventually)

Analysis approach:

  1. Plot weekly/monthly retention cohorts
  2. Find where the curve flattens (that's your "retained" base)
  3. Focus on getting more users past the flattening point

Retention Tactics by Stage

Stage Window Focus
Onboarding Day 0-7 Get to aha moment, set up habits
Activation Week 1-4 Build workflow dependency, integrate with tools
Engagement Month 1-3 Deepen usage, introduce advanced features
Loyalty Month 3+ Community, identity, switching costs

Reducing Churn

  1. Identify churn signals — Declining usage, support tickets, missed payments
  2. Segment churned users — Why did they leave? (Survey, interview, data)
  3. Intervene early — Automated nudges when churn signals appear
  4. Win-back campaigns — Email lapsed users with what's new
  5. Fix the product — Most churn is product churn, not marketing churn

Experimentation (ICE Framework)

Prioritize growth experiments using ICE:

Factor Score 1-10 Definition
Impact How much will this move the metric? 1 = barely, 10 = 2x+
Confidence How sure are we it will work? 1 = wild guess, 10 = proven
Ease How easy to implement and measure? 1 = months, 10 = hours

ICE Score = (Impact + Confidence + Ease) / 3

Experiment Template

## Experiment: {Name}

**Hypothesis:** If we {change}, then {metric} will {improve} because {reason}.

**Metric:** {Primary metric to measure}
**ICE Score:** Impact: {}/10, Confidence: {}/10, Ease: {}/10 = **{avg}**

**Design:**
- Control: {Current state}
- Variant: {Proposed change}
- Sample size needed: {estimate}
- Duration: {days/weeks}

**Results:**
- {Metric}: Control {X} vs. Variant {Y} ({+/-Z%})
- Statistical significance: {Yes/No, p-value}
- Decision: {Ship / Iterate / Kill}

**Learnings:** {What did we learn regardless of outcome?}

Channel Selection Matrix

Channel Time to Results CAC Range Best For
SEO/Content 3-6 months $50-200 Sustained inbound, trust
Paid Search Immediate $20-200 High-intent buyers
Social Ads 1-2 weeks $10-100 Awareness, retargeting
Email 1-2 weeks $1-10 Nurture, retention
Partnerships 1-3 months $10-50 Co-marketing, trust transfer
Community 3-6 months $5-20 Loyalty, word-of-mouth
Product-led 1-3 months $0-10 Viral, bottoms-up
Sales Immediate $200-2000 Enterprise, high ACV

Rule of thumb: Master 1-2 channels before adding more. Most startups spread too thin.

Output Format

# Growth Strategy: {Product/Company}

## North Star Metric
{Metric} — {Why this metric}

## Current Funnel Analysis

| Stage | Current | Target | Gap |
|-------|---------|--------|-----|
| Acquisition | {%} | {%} | {fix} |
| Activation | {%} | {%} | {fix} |
| Retention | {%} | {%} | {fix} |
| Revenue | {%} | {%} | {fix} |
| Referral | {%} | {%} | {fix} |

## Primary Growth Loop
{Description of the main loop to invest in}

## Top 5 Growth Experiments (by ICE Score)

| # | Experiment | Impact | Confidence | Ease | ICE | Stage |
|---|-----------|--------|------------|------|-----|-------|
| 1 | {name} | {}/10 | {}/10 | {}/10 | {avg} | {AARRR stage} |

## 90-Day Roadmap

### Month 1: {Theme}
{Specific actions}

### Month 2: {Theme}
{Specific actions}

### Month 3: {Theme}
{Specific actions}

Important Notes

  • Growth strategy without retention is a leaky bucket. Always fix retention first.
  • Metrics without context are meaningless. Always compare to your own trend line AND industry benchmarks.
  • Most growth comes from compounding small wins, not silver bullets.
  • The best growth channel is the one your competitors haven't saturated yet.
  • Talk to churned users. They'll tell you more than any dashboard.
Weekly Installs
49
GitHub Stars
202
First Seen
Feb 14, 2026
Installed on
claude-code44
opencode43
gemini-cli42
codex39
github-copilot38
cursor37