# Retention Cohorts

> Cohort retention analysis — triangle chart construction, curve shape analysis for PMF detection (flattening/declining/rising), layer cake chart, PMF-driven priority shifting, 5 retention anti-patterns, 4 improvement levers, three definitions setup. Use when measuring PMF, analyzing user retention, or deciding whether to scale or improve product.

- **Type:** Skill
- **Install:** `agentstack add skill-masterleopold-anty-framework-retention-cohorts`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [masterleopold](https://agentstack.voostack.com/s/masterleopold)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [masterleopold](https://github.com/masterleopold)
- **Source:** https://github.com/masterleopold/anty-framework/tree/main/skills/retention-cohorts
- **Website:** https://4n7y.com

## Install

```sh
agentstack add skill-masterleopold-anty-framework-retention-cohorts
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Retention & Cohort Analysis

## When to Apply
- Measuring product-market fit
- Deciding whether to invest in growth vs product improvement
- When the founder asks "do we have PMF?"
- When retention data is available from integrations
- Quarterly PMF reassessment

## Core Framework

### Three Definitions (Set During Onboarding)

Before any retention measurement, guide the founder to define:

1. **Cohort grouping** — How new users are grouped
   - Weekly: daily-use products
   - Monthly: utility products
   - Quarterly: infrequent-use products (travel, tax)

2. **Active action** — What counts as "active." Must reflect genuine value delivery.
   - Ask: "Imagine watching a customer use your product. What moment tells you they're genuinely getting value?"
   - B2B SaaS: "completed a core workflow"
   - Consumer: "engaged with 3+ pieces of content"
   - Marketplace: "completed a transaction"

3. **Time granularity** — How often users should ideally use the product. Cross-check against chosen action for consistency.

### Triangle Chart (Cohort Retention Table)

```
         Week 0  Week 1  Week 2  Week 3  Week 4  Week 5
Jan      100%    62%     45%     38%     35%     34%    number." |

### 4 Improvement Levers

When retention curves don't flatten:

1. **Product improvement** — New use cases, speed, simpler flows
2. **Better user acquisition** — Targeting users who are a better fit. "Paid cohorts retain worse than organic. Consider shifting budget."
3. **Onboarding/activation** — Help users reach "aha moment" faster. Often cheapest lever. "What was the user doing yesterday? What should they do differently today?"
4. **Network effects** — If applicable, user-to-user value. "Each new user could make your product better for existing users."

## Decision Rules

1. **Three definitions before measurement** — no retention analysis without explicit cohort, action, and time definitions
2. **Shape over numbers** — 20% that flattens beats 50% that declines
3. **PMF drives priorities** — no scaling without retention flattening
4. **Newer vs older cohorts** — compare across time to detect product trajectory
5. **Layer cake for growth truth** — reveals treadmill vs genuine growth
6. **Cheapest lever first** — onboarding improvement often has highest ROI

## Anti-Patterns to Detect

| Anti-Pattern | Signal | Response |
|---|---|---|
| Scaling before PMF | Growing acquisition with declining retention | "Retention curve hasn't flattened. Fix retention before scaling." |
| Wrong granularity | Mismatched time period and product usage | "Adjust cohort grouping to match actual product usage frequency." |
| Vanity retention | Tracking logins instead of value delivery | "Redefine 'active' to reflect genuine value delivery." |
| Single-number fixation | "Our retention is 40%" | "40% when? Show the full curve over time." |

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [masterleopold](https://github.com/masterleopold)
- **Source:** [masterleopold/anty-framework](https://github.com/masterleopold/anty-framework)
- **License:** MIT
- **Homepage:** https://4n7y.com

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-masterleopold-anty-framework-retention-cohorts
- Seller: https://agentstack.voostack.com/s/masterleopold
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.