Growth Loops

---
name: growth-loops
description: "Maps acquisition, engagement, and monetization loops to identify self-reinforcing growth mechanisms. Moves beyond the linear funnel model to identify compounding loops where output from one cycle becomes input for the next. Based on Reforge's growth loops framework."
category: gtm
default_depth: standard
---

# Growth Loops Identification

Read `docs/workflow-contract.md` once per session before applying this skill. Resolve it from the nearest ancestor of this file containing `manifest.json`; all Shipwright paths are relative to that root.

## Description

Maps acquisition, engagement, and monetization loops to identify self-reinforcing growth mechanisms. Moves beyond the linear funnel model to identify compounding loops where output from one cycle becomes input for the next. Based on Reforge's growth loops framework.

## When to Use

- Building or refining a growth strategy
- Diagnosing why growth has plateaued
- Evaluating which growth investments will compound vs. be one-time
- Aligning product, marketing, and engineering on growth priorities

## Key Concepts

**Growth loop vs. funnel:** A funnel is linear (top → bottom, then done). A loop is circular (output feeds back as input, creating compounding growth).

**Loop components:**
1. **Input**, what enters the loop (new users, content, data)
2. **Action**, what the user does inside the product
3. **Output**, what the action produces that feeds back as new input

## Depth

| Scope | Use When | Sections to Include |
|---|---|---|
| **Light** | Quick audit of what loops exist today | Step 1: Identify Existing Loops, Step 2: Evaluate Loop Strength |
| **Standard** | Growth strategy planning or plateau diagnosis | All sections (Steps 1-4) |
| **Deep** | Full growth model redesign or investor-facing growth narrative | All sections + loop interaction map, multi-quarter compounding projections, sensitivity analysis on key assumptions |

**Omit rules:** At Light depth, skip Step 3 (Intervention Opportunities) and Step 4 (Design New Loops). Produce only the Loop Inventory and Evaluation table.

## Framework

### Step 1: Identify Existing Loops

Map every loop currently operating in the product:

```markdown
## Acquisition Loops (how new users arrive)

### Loop: [Name, e.g., "User-Generated Content SEO"]
```
New user → Creates content → Content indexed by Google → New users discover via search → [repeat]
```

**Input:** [What starts the cycle]
**Action:** [What the user does]
**Output:** [What this produces that feeds back]
**Cycle time:** [How long one loop takes]
**Current performance:** [Metrics, e.g., "Each piece of content generates 12 visits/month"]
**Compounding?** [Yes/No, does it get stronger over time?]

### Loop: [Name, e.g., "Viral Invite"]
```
User gets value → Invites colleague → Colleague signs up → Gets value → Invites more → [repeat]
```

**Viral coefficient:** [K-factor, invites × conversion rate]
**Current K:** [number, >1 means viral growth, <1 means loop decays]

---

## Engagement Loops (how users come back)

### Loop: [Name, e.g., "Data Network Effect"]
```
User adds data → Product gets smarter → Better recommendations → More usage → More data → [repeat]
```

**Retention driver:** [What specifically brings users back]
**Frequency:** [How often this loop fires, daily, weekly, monthly]
**Strength indicator:** [Metric, e.g., "Users who complete X retain at 2x the baseline"]

---

## Monetization Loops (how revenue grows)

### Loop: [Name, e.g., "Seat Expansion"]
```
Team lead adopts → Team sees value → More seats purchased → More usage data → Proves ROI → Budget increase → [repeat]
```

**Revenue impact:** [How this loop contributes to revenue growth]
**Net Revenue Retention:** [NRR %, >100% means expansion outpaces churn]
```

### Step 2: Evaluate Loop Strength

```markdown
## Loop Evaluation

| Loop | Type | Cycle Time | Compounding? | Current Strength | Investment Level |
|---|---|---|---|---|---|
| [Loop 1] | Acquisition | [time] | Yes/No | Strong/Medium/Weak | High/Med/Low |
| [Loop 2] | Engagement | [time] | Yes/No | Strong/Medium/Weak | High/Med/Low |
| [Loop 3] | Monetization | [time] | Yes/No | Strong/Medium/Weak | High/Med/Low |

### Strength Assessment Criteria:
- **Strong:** Loop is self-sustaining, growing without proportional investment
- **Medium:** Loop operates but requires ongoing investment to maintain
- **Weak:** Loop exists in theory but isn't producing meaningful results yet
```

### Step 3: Identify Highest-Leverage Interventions

```markdown
## Intervention Opportunities

For each loop, identify where a small improvement would have the biggest compounding effect:

### Loop: [Name]
**Bottleneck:** [Where the loop loses the most volume]
**Current conversion at bottleneck:** [X]%
**Intervention:** [What we could do to improve it]
**Estimated impact:** [If we improve bottleneck from X% to Y%, loop output increases by Z%]
**Effort:** [T-shirt size]
**Compounding effect:** [How this improvement multiplies over N cycles]
```

### Step 4: Design New Loops

If existing loops are insufficient, design new ones:

```markdown
## Proposed New Loop: [Name]

**Thesis:** We believe that [mechanism] will create a self-reinforcing loop because [reasoning].

**Loop diagram:**
[Step 1] → [Step 2] → [Step 3] → feeds back to [Step 1]

**Key assumptions:**
1. [Assumption, e.g., "Users will share because they get value from sharing"]
2. [Assumption, e.g., "Shared content will convert at >5%"]

**MVP test:** [Smallest thing we can build to test if this loop works]
**Success criteria:** [What metric at what level validates the loop]
```

## Minimum Evidence Bar

**Required inputs:** Product description, current growth channels, and at least one of: usage/analytics data, cohort retention curves, or user acquisition source breakdown.

**Acceptable evidence:** Product analytics (cycle time, conversion rates at each loop step), cohort retention data, referral/invite metrics, NRR figures, A/B test results on loop interventions.

**Insufficient evidence:** If no product analytics or retention data exist, stop and recommend running the **metrics-dashboard** skill to instrument core loop steps before attempting this skill.

**Hypotheses vs. findings:**
- **Findings:** Loop Inventory (which loops exist) and Evaluation ratings (Strong/Medium/Weak) must be grounded in observed metrics or user behavior.
- **Hypotheses:** New Loop Proposals and compounding projections are inherently speculative, must be labeled "Hypothesis, validate with MVP test."

## Output Format

Produce a Growth Loops Analysis with:
1. **Loop Inventory**, all currently operating loops mapped
2. **Loop Evaluation**, strength, cycle time, compounding assessment
3. **Intervention Priorities**, highest-leverage improvements ranked
4. **New Loop Proposals**, designed loops with hypotheses and tests
5. **Investment Recommendation**, where to focus growth engineering effort

**Shipwright Signature (required closing):**
6. **Decision Frame**, recommended loop to invest in first, trade-off vs. alternative loops, confidence level with data quality, owner (growth or product lead), decision date, revisit trigger (loop metric review at end of cycle)
7. **Unknowns & Evidence Gaps**, unmeasured cycle times, assumed conversion rates, missing K-factor data
8. **Pass/Fail Readiness**, PASS if at least one loop has measured cycle time and conversion data at each step; FAIL if all loops are theoretical with no observed metrics
9. **Recommended Next Artifact**, Which Shipwright skill to run next and why

## Common Mistakes to Avoid

- **Funnel thinking**, Funnels are one-and-done; loops compound. Design for loops.
- **Ignoring cycle time**, A strong loop that takes 6 months per cycle won't save you this quarter
- **Over-investing in weak loops**, If the loop isn't working after reasonable investment, try a different loop
- **All acquisition, no engagement**, Leaky bucket. Fix engagement loops before pouring more into acquisition.
- **Assuming virality**, Most products aren't viral (K < 1). That's fine. Content loops, paid loops, and sales loops compound too.

## Weak vs. Strong Output

**Weak:**
> "We have a viral loop where users invite other users, which drives growth."

No K-factor, no cycle time, no conversion rate, impossible to evaluate or invest against.

**Strong:**
> "Viral Invite loop: K-factor 0.3 (avg 2.1 invites/user x 14% acceptance rate), 11-day cycle time. Loop is decaying (K < 1) but intervention on invite acceptance (current 14% → target 25% via pre-filled workspace) would lift K to 0.53 and reduce payback period by 4 months."

Measured, actionable, with a specific intervention tied to a projected outcome.