game-economy-and-monetization

Game economy- and monetization-design skill

You are producing an economy/monetization artifact — an economy spreadsheet, a gacha-math sheet, or a retention-KPI dashboard. The engineering-side game-design economy template in pair-programmer covers the basic tabs; this skill adds the Monte-Carlo balance pass, the full per-region jurisdiction matrix, and the live-service KPI dashboard. The deliverable is a balanced, simulated, legally-mapped economy spec — never code, never the live store.

This skill feeds the loot-box-jurisdiction gate (HITL, owned jointly by The Economist and The Arbiter). Any randomized monetization that cannot fill the jurisdiction table does not pass.

The cardinal rule

If a paid mechanic has random outcomes, the odds are published, there is a pity floor, and the jurisdiction table is filled for every shipping region — or it does not ship. Belgium and the Netherlands restrict/ban loot boxes; you disable or restructure for them, you do not "risk it".

Anti-patterns to refuse

• Undisclosed odds — any randomized paid mechanic without published drop rates. Violates Apple/Google policy, Chinese/Korean/Japanese law, and ESRB notice. Non-negotiable.
• No pity floor — a gacha with no guaranteed-by-N counter reads as predatory and erodes player trust. A hard pity is a strong design + consumer-protection default (and some markets, e.g. China, regulate probability/disclosure) — but treat "no pity" as a design/trust red flag, not a blanket legal failure. Cite the specific market rule when you claim a legal requirement.
• Ignoring BE/NL — shipping paid randomized rewards to Belgium (enforcement- restricted; publishers commonly disable) or the Netherlands (contested/fact- specific after the 2022 Council of State reversal) without a region-aware plan. Decide per-region with counsel; an account-region gate is the usual mechanism.
• Kompu-gacha — "complete-the-set to win a rare" mechanics are illegal in Japan. Refuse the pattern entirely, not just for JP.
• Pay-to-win with no skill ceiling — purchasable power with no skill expression. Beyond the ethics, it kills retention and triggers AU/ratings scrutiny. Monetize cosmetics, time, and breadth — not the win condition.
• Number-soup economy — currencies with no source/sink/leak accounting; the faucet/drain MUST balance or the economy inflates/starves.
• KPIs with no balance simulation — projecting LTV/ARPDAU off a spreadsheet point estimate instead of a Monte-Carlo distribution.

Templates

economy_spreadsheet template (tabs)

# Economy spreadsheet — <Title>

## Currencies (tab)
| Currency | Premium? | Source(s) | Sink(s) | Leak(s) | Daily faucet | Daily drain | Net | Per-region note |
| Gold     | no       | quests, drops | gear, repair | decay | 5,000 | 4,600 | +400 | — |
| Gems     | yes      | IAP, events   | gacha, skips | — | — | — | — | see jurisdiction tab |

## Drop tables (tab)
| Source | Item | Rarity | Weight | Drop % (PUBLISHED) | Pity timer | Floor rate |
| Banner A | 5★ char | SSR | 0.6  | 0.60% | hard pity @ 90 | guaranteed by 90 |
| Banner A | 4★      | SR  | 5.1  | 5.10% | soft pity @ 75 | — |

## Per-region jurisdiction (tab)
| Region | Loot box | Odds published | Age gate | Special rule | Action |
| Belgium (BE) | BANNED | n/a | n/a | gambling-law ruling | DISABLE paid random; sell direct |
| Netherlands (NL) | RESTRICTED | yes | yes | tradeable-reward scrutiny | non-tradeable + odds, or disable |
| China (CN) | allowed | MANDATORY publish | yes | publish exact rates; anti-addiction time limits | publish + minor caps |
| South Korea (KR) | allowed | MANDATORY publish (2024 law) | yes | GMC oversight | publish exact rates |
| Japan (JP) | allowed | publish (industry) | — | NO kompu-gacha (illegal) | publish; ban complete-gacha |
| Apple (iOS, global) | allowed | REQUIRED (App Store 3.1.1) | yes | must disclose odds | publish in-app |
| Google (Play, global) | allowed | REQUIRED (Play policy) | yes | must disclose odds | publish in-app |
| Australia (AU) | allowed | — | M-rating | sim-gambling content → M+ | rating note → Arbiter |
| EU general | watch | yes | 18+ likely | DFA / consumer-law pressure | publish + age-gate, monitor |
| US (ESRB) | allowed | ESRB "in-game purchases (random)" label | yes | label required | apply label |

## Progression curves (tab)
| Level | XP req | Cumulative | Power | Time-to-level (median sessions) | Paywall pressure |

## Balance matrix (tab) — Monte-Carlo outputs
| Cohort | Sim runs | Median sessions-to-paywall | P5 | P95 | Inflation @ day 30 |
| F2P    | 100,000  | 18 | 9  | 41 | +2.1% |

gacha-math sheet template

# Gacha-math sheet — <Banner>

## Published rates (must match in-game disclosure exactly)
| Tier | Rate | Pity |
| SSR  | 0.60% | hard pity @ 90, soft pity ramp from 75 |
| SR   | 5.10% | guaranteed ≥1 per 10-pull |

## Pity model
<Soft pity: SSR rate ramps +6%/pull from pull 75. Hard pity: SSR @ 90.
50/50 rate-up rule stated. Carry-over across banners: yes/no.>

## Monte-Carlo expected cost
| Outcome | Median pulls | P95 pulls | Median $ | P95 $ | Region caps |
| 1× SSR  | 62           | 90        | $X       | $Y    | CN minor cap applies |

## Floor guarantee
<The worst-case spend a player can hit — this is the consumer-protection number.>

## Kompu-gacha check
<CONFIRM: no complete-the-set-to-unlock mechanic exists (illegal in JP).>

retention-KPI dashboard template

# Retention-KPI dashboard — <Title>

| KPI | Definition | Target | Alert floor | Source |
| D1 retention | % returning day 1 | ≥40% | <30% | telemetry |
| D7 / D30     | % returning       | ≥20% / ≥10% | <12% / <5% | telemetry |
| ARPDAU       | revenue / DAU     | $0.15 | <$0.08 | store |
| LTV (180d)   | Monte-Carlo dist. | $4.20 median | P5 < $1.50 | sim |
| Conversion   | % payers          | ≥3% | <1.5% | store |
| Whale concentration | % rev from top 1% | <50% | >70% (risk) | store |

## Store A/B plan
| Test | Variant A | Variant B | Metric | Min sample | Stop rule |
| price point | $4.99 | $5.99 | rev/install | 10k/arm | 95% sig or 14d |
<A/B changes to live economy are HITL (RLM-GAMING §8) → The Custodian.>

Constraints

• Randomized paid mechanics MUST publish odds, include a hard pity floor, and fill the jurisdiction table for EVERY shipping region — or they do not ship (loot-box-jurisdiction, HITL).
• BE = disable paid random; NL = restructure or disable; JP = no kompu-gacha (refuse the pattern globally); CN/KR/JP/Apple/Google = publish exact odds; AU = rating note to The Arbiter.
• No pay-to-win without a preserved skill ceiling; monetize cosmetics, time, and breadth — never the win condition.
• Every currency MUST balance source/sink/leak; project LTV/ARPDAU from a Monte-Carlo distribution (≥100k runs), never a point estimate.
• Every economy artifact carries a per-region note (this skill's core duty), an accessibility note (spending-control / parental-control surfaces; cross-ref game-accessibility-guidelines), and — for online economies — a server-authority note (server owns wallets, drops, and transactions; cross-ref server-authority-fairplay).
• Live economy changes are HITL via The Custodian; store A/B follows a stated stop rule.
• Cite reference titles for monetization model precedent (e.g. Genshin Impact pity model, Marvel Snap pity, Path of Exile cosmetic-only F2P).
• Produces only design/spec markdown and DECISION_RECORDs. Store integration, payment code, and telemetry pipelines are a DEV_TASK to engineering; store art is an ASSET_JOB to garland. Never build the store here. See game-studio-pipeline.