Design a product recommendation system for my online store

Help me design a personalized product recommendation system for my online store. Store and catalog: [WHAT WE SELL, number of products, how often the catalog changes] Traffic and customers: [monthly visitors, orders, share of logged-in or returning customers] Data available: [browsing events, purchases, cart adds, ratings, product attributes, search queries] Platform and team: [e.g. Shopify, custom stack; in-house data skills or none] Goal: [higher average order value / conversion / repeat purchases / discovery] Deliver: 1. Approach by maturity: start with what works for my data size, e.g. rules and bestsellers, "frequently bought together" from co-purchases, content-based similarity from product attributes, collaborative filtering, then learned ranking models. Say which stage fits me now and when to move up. 2. Placements: product page, cart, homepage, email, search, post-purchase, with the best recommendation type for each. 3. Cold start: how to handle new visitors and new products. 4. Business rules: stock and margin filters, diversity, avoiding recommending items just bought, promoted items, and fairness to smaller brands if relevant. 5. Build vs. buy: platform apps or vendors vs. a custom model, with trade-offs. 6. Measurement: offline metrics (precision, recall, coverage) and an A/B test plan with the business metric that decides success. 7. Privacy: consent for tracking, data retention, and clear explanations to customers. If I want to build it myself, outline the data pipeline and give starter code in [Python/SQL].