Design the API and data model considerations for quick commerce replenishment

Problem Statement Description

Zepto wants to improve replenishment for busy households who repeatedly buy essential items such as milk, bread, eggs, fruits, snacks, baby products, and cleaning supplies. The product experience needs to help customers complete baskets quickly while ensuring that the backend can reliably understand replenishment intent, check local dark-store availability, handle substitutions, and avoid recommending unavailable or stale inventory.

Your task is to define the API and data model considerations for a quick-commerce replenishment capability. Focus on what systems need to exchange, what entities and relationships must be represented, how customer and inventory data should be handled, and how the experience can remain reliable under hyperlocal, high-frequency ordering conditions.

Assume the feature may support use cases such as “buy again,” smart reminders, household staples lists, subscription-like repeat purchases, and replenishment questions during cart building. You do not need to design the UI in detail, but you should account for how product requirements translate into backend capabilities, data contracts, and operational constraints.

The experience should consider:

- Core users and workflows, including busy households, repeat buyers, pickers, inventory systems, and fulfillment operations

- API requirements for replenishment suggestions, cart additions, availability checks, substitutions, pricing, delivery promise, and order confirmation

- Data model entities such as customer, household, SKU, product variant, store, inventory, purchase history, replenishment signal, cart, order, and substitution preferences

- Data quality challenges, including SKU mapping, out-of-stock items, stale inventory, picker accuracy, and changing pack sizes or prices

- Reliability and latency expectations in a quick-commerce environment where availability and delivery promises can change quickly

- Privacy, consent, and security considerations for using purchase history, household patterns, addresses, and personalization signals

- Observability, instrumentation, and error handling needed to monitor API health, recommendation quality, stockout impact, and basket completion

- Rollout and product trade-offs, including operational load, fallback behavior, accessibility, experimentation, and responsible use of automation

The goal is to demonstrate how you would translate a replenishment product need into clear technical requirements and data considerations, while balancing customer trust, basket completion, dark-store realities, and scalable system design.