Referral Program

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

.codex/skills/a11y-audit

.codex/skills/ab-test-setup

.codex/skills/account-executive

.codex/skills/ad-creative

.codex/skills/aeo

Audit 0.4.1 · source sha 6e3fcbdbb5cf · 19 Oct 2026Report an issue or request a re-scan

• Publisherborghei GitHub userOwner of the repository this skill is read from
• Source folderborghei/Claude-Skills/business-growth/referral-programThe exact folder we audited
• LicenceNOASSERTIONAs the repository states it
• Commit pinned for this verdict6e3fcbdbb5cfA new audit runs when the source changes
• Repository statusNot archivedLast push 12 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Referral and affiliate program design covering referral loop architecture, incentive design, trigger moment optimization, viral coefficient modeling, affiliate program structure, and optimization playbook.

---
name: referral-program
description: >
  Referral and affiliate program design covering referral loop architecture,
  incentive design, trigger moment optimization, viral coefficient modeling,
  affiliate program structure, and optimization playbook.
license: MIT + Commons Clause
metadata:
  version: 1.1.0
  author: borghei
  category: business-growth
  updated: 2026-06-15
  tags: [referral, affiliate, growth, viral, word-of-mouth, acquisition]
---
# Referral Program

Production-grade referral and affiliate program framework covering the 4-stage referral loop, incentive design methodology, trigger moment optimization, share mechanics, viral coefficient modeling, affiliate program architecture, and systematic optimization playbook. Designed to build programs that compound, not collect dust.

## Core Capabilities

- **Program type & loop design** — referral vs affiliate decision, plus the 4-stage loop (trigger → share → convert → reward)
- **Incentive design** — single- vs double-sided, reward types, tiered gamification, reward economics against LTV/CAC
- **Trigger & share mechanics** — in-product and email trigger points, share channel priority, first-person share copy
- **Referred-user experience** — referral landing page, attribution rules, program copy set (prompts, emails, dashboards)
- **Growth math** — K-factor modeling, revenue impact models, and lever-by-lever K improvement
- **Affiliate framework** — commission models, tier systems, partner toolkit, recruitment
- **Optimization** — diagnose-before-optimize playbook, metric benchmarks, troubleshooting, and three Python tools

## When to Use

- The user asks to "design a referral program", "launch an affiliate program", or "improve viral growth"
- The decision between customer referral vs affiliate program needs to be made
- An existing referral program has stalled (K-factor <1, low share rate, low referred-user conversion)
- Reward structure needs sizing against CAC, margin, or LTV
- Trigger moments need to be identified (when to ask, which in-product events, which lifecycle emails)
- The user says "word-of-mouth isn't working" or "we want to add a refer-a-friend flow"

## Clarify First

Before designing the referral program, confirm these inputs. If any is unknown or vague, ASK — do not assume:

- [ ] **Program type** — customer referral vs affiliate (enthusiastic/social customers vs team buyers) (selects the entire framework)
- [ ] **Trigger moment** — the in-product or lifecycle point where you ask (a broken Stage 1 can't be fixed by a bigger reward at Stage 4)
- [ ] **Reward economics** — first-payment value, margin, and CAC (caps the reward at <30% of first payment)
- [ ] **Current referral rate (if any)** — decides single- vs double-sided incentive and which stage to fix first

Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the deliverable.

## Quick Start

1. **Pick the program type** — use the Referral vs Affiliate Decision table (enthusiastic/social customers → referral; team buyers → affiliate).
2. **Build the loop in order** — trigger → share → convert → reward; a broken Stage 1 can't be fixed by a bigger reward at Stage 4.
3. **Size the incentive** — cap reward at <30% of first payment; go double-sided if referral rate <1%.
4. **Model and validate** — run the scripts (`referral_economics_calculator.py`, `referral_funnel_analyzer.py`, `affiliate_commission_modeler.py`) to size rewards, find the weakest stage, and model affiliate tiers.
5. **Optimize by priority** — fix awareness first, then share flow, then referred experience, then the incentive.

## References

Load the reference that matches the task — keep this file lean and pull detail on demand:

- **[references/loop-and-incentives.md](references/loop-and-incentives.md)** — Referral vs Affiliate decision table, the full 4-stage loop with per-stage tables, incentive design (single/double-sided, reward types, tiers, economics), and trigger moment architecture. Read when designing the core program.
- **[references/share-and-experience.md](references/share-and-experience.md)** — share channel priority, share message templates, referral landing page layout, attribution rules, and the program copy set (in-app prompt, dashboard, post-activation email). Read when building the sharing flow and referred-user experience.
- **[references/modeling-and-affiliate.md](references/modeling-and-affiliate.md)** — K-factor calculation and improvement levers, plus the full affiliate framework (commission structure, tier system, toolkit, recruitment). Read when modeling growth math or designing an affiliate program.
- **[references/optimization-and-operations.md](references/optimization-and-operations.md)** — optimization playbook, key metrics and benchmarks, revenue impact model, output artifacts, full tool reference, troubleshooting table, success criteria, and anti-patterns. Read when diagnosing a stalled program or operating the scripts.

## Scope & Limitations

**In scope:** Customer referral program design (4-stage loop), incentive structure (single-sided, double-sided, tiered), trigger moment architecture, share mechanics, referral landing page specifications, viral coefficient modeling, affiliate program framework (commission models, tier systems, recruitment), and systematic optimization playbook.

**Out of scope:** Referral landing page visual design and CRO (use page-cro), signup flow optimization for referred users (use signup-flow-cro), post-signup onboarding for referred users (use onboarding-cro), churn prevention for referred customers (use churn-prevention), and reward pricing alignment (use pricing-strategy). Scripts operate on local data only -- no integrations with referral platforms (ReferralHero, Viral Loops, PartnerStack, etc.).

**Limitations:** K-factor benchmarks assume consumer or prosumer SaaS; B2B enterprise referral programs have different dynamics (lower K but hi

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

git clone https://github.com/borghei/Claude-Skills
git -C Claude-Skills checkout 6e3fcbdbb5cf

Works in

1. 2026-10-19
   !CAUTION B 89 / 100

   first audit · source 6e3fcbdbb5cf

Questions about Referral Program

• github.com/borghei/Claude-Skills/tree/main/business-growth/refer — the audited source at commit 6e3fcbdbb5cf and every finding aboveObserved · 19 Oct 2026

Spotted something wrong? Report a correction · Methodology