petrogurcak/growth-hacking

Growth Hacking

Overview

This skill provides frameworks for systematic, data-driven growth using methodologies from Sean Ellis & Morgan Brown's "Hacking Growth". Use when you need to prioritize experiments and drive sustainable growth.

Announce: "I'm using growth-hacking to set up systematic experimentation for growth."

When to Use

USE this skill:

• "How do I grow faster?"
• "Which marketing channel should I try first?"
• "Do I have product-market fit?"
• Prioritizing growth experiments
• Setting up growth process
• Measuring if product is "must-have"

DON'T use this skill:

• Creating offers → offer-creation
• Positioning → uvp-optimization
• Product onboarding → product-led-growth
• SaaS metrics → saas-bootstrap

Iron Law

NO GROWTH WITHOUT PRODUCT-MARKET FIT

Don't pour water into a leaky bucket.
First validate PMF with Must-Have Survey.
If <40% "very disappointed" → fix product first.

Core Frameworks

Framework 1: Must-Have Survey (PMF Validation)

Purpose: Determine if you have Product-Market Fit before scaling.

The Question:

"How would you feel if you could no longer use [product]?"

• Very disappointed
• Somewhat disappointed
• Not disappointed
• N/A (no longer use it)

Interpretation: | Very Disappointed % | PMF Status | Action | |--------------------|------------|--------| | 40%+ | PMF achieved | Scale growth | | 25-40% | Almost there | Improve core value | | <25% | No PMF | Pivot or major changes |

When to use:

• Before scaling marketing spend
• Before hiring growth team
• When unsure if product resonates
• Quarterly health check

Additional questions to ask:

1. "What type of person would benefit most from [product]?"
2. "What is the main benefit you receive from [product]?"
3. "How can we improve [product] for you?"

---

Framework 2: Growth Hacking Process

Purpose: Systematic cycle for continuous growth experimentation.

The Cycle:

    ┌─────────────┐
    │   ANALYZE   │ ← Data collection, insights
    └──────┬──────┘
           │
           ▼
    ┌─────────────┐
    │   IDEATE    │ ← Generate experiment ideas
    └──────┬──────┘
           │
           ▼
    ┌─────────────┐
    │ PRIORITIZE  │ ← Score with ICE
    └──────┬──────┘
           │
           ▼
    ┌─────────────┐
    │    TEST     │ ← Run experiments
    └──────┬──────┘
           │
           └──────────→ Back to ANALYZE

Phase Details:

1. ANALYZE

• What do the numbers say?
• Where is the biggest drop-off?
• What patterns emerge from user behavior?
• What qualitative feedback exists?

2. IDEATE

• Brainstorm without judgment
• Quantity over quality initially
• Cross-functional input (product, marketing, sales, support)
• Document every idea in backlog

3. PRIORITIZE

• Use ICE scoring (see Framework 3)
• Weekly prioritization meeting
• Select top 3-5 experiments for the week

4. TEST

• Define hypothesis clearly
• Set success metrics upfront
• Run for statistical significance
• Document learnings regardless of outcome

---

Framework 3: ICE Scoring

Purpose: Objectively prioritize experiment ideas.

Formula:

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

Scoring (1-10 scale):

| Dimension | What it measures | Scoring Guide | | -------------- | ----------------------------- | ------------------------------------- | | Impact | Expected effect on key metric | 10 = 10x improvement, 1 = tiny effect | | Confidence | How sure are we it'll work? | 10 = data-backed, 1 = pure guess | | Ease | Resources/time to implement | 10 = <1 day, 1 = months of work |

Example Scoring:

| Experiment | Impact | Confidence | Ease | ICE | | --------------------- | ------ | ---------- | ---- | --- | | New headline A/B test | 7 | 6 | 9 | 7.3 | | Complete redesign | 9 | 4 | 2 | 5.0 | | Add testimonials | 5 | 8 | 8 | 7.0 | | Build referral system | 8 | 5 | 3 | 5.3 |

Priority: Headline test (7.3) > Testimonials (7.0) > Referral (5.3) > Redesign (5.0)

---

Framework 4: North Star Metric

Purpose: Single metric that captures core value delivery.

Characteristics of good North Star:

• Measures value delivered to customers
• Leading indicator of revenue
• Everyone can influence it
• Simple to understand

Examples by business type:

| Business Type | North Star Metric | | ------------- | ----------------------- | | SaaS | Weekly Active Users | | Marketplace | Transactions completed | | E-commerce | Purchases per month | | Media | Time spent reading | | Freemium | Free-to-paid conversion |

---

Workflow

Phase 1: PMF Check

Before any growth work:

1. Run Must-Have Survey
2. If <40% "very disappointed" → STOP growth, fix product
3. If 40%+ → proceed to growth process

Phase 2: Setup Growth Process

Weekly cadence:

• Monday: Analyze previous week's experiments
• Tuesday: Ideation session
• Wednesday: ICE scoring and prioritization
• Thursday-Friday: Implementation and launch

Phase 3: Run Experiments

For each experiment:

## Experiment: [Name]

**Hypothesis:** We believe [change] will [effect] because [reasoning].

**Metric:** [What we're measuring]

**Success criteria:** [X]% improvement with 95% confidence

**Duration:** [X] days/weeks

**Result:** [Outcome + learnings]

Phase 4: Learn and Iterate

• Document ALL learnings (wins and failures)
• Share across team
• Feed insights back into Analyze phase
• Update experiment backlog

---

Quick Router

| Situation | Framework to Use | | ----------------------------------- | -------------------------------------------- | | "Should we scale marketing?" | Must-Have Survey first | | "Too many ideas, can't decide" | ICE Scoring | | "Growth is random and chaotic" | Growth Process (weekly cadence) | | "What metric should we focus on?" | North Star Metric | | "Our experiments don't move needle" | Check if targeting right metric (North Star) |

---

Growth Team Structure

Minimum viable growth team:

• Growth Lead (owns process)
• Engineer (implements)
• Data Analyst (measures)
• Designer (optional but helpful)

Cross-functional input from:

• Marketing
• Product
• Customer Support
• Sales

---

Templates

ICE Scoring Sheet

| #   | Experiment Idea | Impact (1-10) | Confidence (1-10) | Ease (1-10) | ICE Score | Status |
| --- | --------------- | ------------- | ----------------- | ----------- | --------- | ------ |
| 1   |                 |               |                   |             |           |        |
| 2   |                 |               |                   |             |           |        |
| 3   |                 |               |                   |             |           |        |

Experiment Brief

## Experiment: [Name]

**Date:** [Start date]
**Owner:** [Who's responsible]

### Hypothesis

We believe that [specific change]
will cause [expected effect]
because [reasoning/data].

### Metrics

- Primary: [Main metric to move]
- Secondary: [Other metrics to watch]

### Success Criteria

- Minimum: [X]% improvement
- Target: [Y]% improvement
- Duration: [X] days minimum

### Implementation

- Changes needed: [List]
- Effort estimate: [Hours/days]

### Results — Dual Threshold Evaluation

- Actual lift: [X]%
- Statistical significance: p=[value] (need p < 0.05)
- Practical significance: lift >= 15%? [Yes/No]
- Verdikt: [WINNER / MARGINAL / TRENDING / NO EFFECT]
- Key learning: [What we learned]

**Verdikt pravidla:**

| Stat. signif. (p < 0.05) | Lift >= 15% | Verdikt       | Akce                                               |
| ------------------------ | ----------- | ------------- | -------------------------------------------------- |
| Ano                      | Ano         | **WINNER**    | Implementuj natrvalo.                              |
| Ano                      | Ne          | **MARGINAL**  | Statisticky real, ale nestoji za effort. Archivuj. |
| Ne                       | Ano         | **TRENDING**  | Slibne, ale malo dat. Pokracuj v experimentu.      |
| Ne                       | Ne          | **NO EFFECT** | Zastav. Presmeruj effort jinam.                    |

Ne jen "je to lepsi" — musí to být DOST lepsi na to, aby se vyplatila implementace.

---

Common Mistakes

Skipping PMF validation: "Let's scale ads!" with no PMF data → Burning money.

Over-optimizing for wrong metric: Increasing signups while ignoring activation → Vanity metrics.

Not running experiments long enough: Stopping at 2 days → No statistical significance.

Declaring winner too early: "Varianta B má +5% po 3 dnech!" → Potřebuješ p < 0.05 AND lift >= 15%. Jinak je to šum.

HiPPO decisions: "CEO thinks we should..." → Use ICE scoring objectively.

One experiment at a time: Serial testing is slow → Run parallel experiments on different funnels.

---

Red Flags

STOP and reassess if:

• Must-Have Survey <25% → Product problem, not growth problem
• All experiments failing → Wrong hypothesis, wrong metric
• No clear North Star → Align team first
• No data infrastructure → Can't measure = can't grow

---

Integration

Before growth-hacking:

• uvp-optimization → Clear positioning
• offer-creation → Compelling offer

After growth experiments identify winners:

• web-copy → Scale winning copy
• product-led-growth → Optimize onboarding
• ux-optimization → Implement UX improvements

---

Cadence Checklist

Weekly:

• [ ] Review last week's experiment results
• [ ] Update ICE scores with new data
• [ ] Select top experiments for this week
• [ ] Launch new experiments

Monthly:

• [ ] Review North Star metric trend
• [ ] Analyze experiment win rate
• [ ] Update growth backlog
• [ ] Share learnings with company

Quarterly:

• [ ] Re-run Must-Have Survey
• [ ] Review if North Star still relevant
• [ ] Major growth strategy review