Personalized Store & Offer Recommendation

You are a monetization data scientist building a personalized in-game store and offer recommendation system: Game type: [INSERT - mobile RPG, shooter, cosmetics-driven, subscription] Catalog: [INSERT - SKUs, bundles, passes] Regions and stores: [INSERT - iOS, Android, web, regional pricing] Ethics posture: [INSERT - no dark patterns, spending controls] Cover the full pipeline: 1. Design - offer strategy and value framework, player-first pricing rules, caps and cooldowns, compliance with platform and regional regulations. 2. Player modeling - preference and taste models, willingness-to-pay and price sensitivity, lifecycle and churn features, whale vs spender protection. 3. Recommendation - contextual bandit or ranking over offers, bundle assembly, discount and timing optimization, exploration of new content. 4. Technology - real-time feature serving, store config and pricing service, storefront rendering, receipt validation, fallback to curated defaults. 5. Data - incremental ARPDAU via holdouts, conversion and refund rates, cannibalization, long-term LTV and retention, player satisfaction. 6. Testing - randomized price experiments, elasticity measurement, regional fairness audits, spending-harm and age-protection tests. 7. Launch - curated baseline, rollout by region, customer support and refund readiness, transparent value communication.