Prove the need for custom work
List the requirements that a configurable store or existing extension cannot satisfy. Separate a product catalogue, a quote workflow and a paid checkout. They create different validation, permission and support needs. Keep an existing system when the actual problem can be solved through product-data cleanup or configuration.
If your products and order flow fit a standard store, compare the ecommerce website package before scoping a custom Django build.
The store requirements checklist helps define the buying journey before choosing the implementation.
Model products, variants and fitment deliberately
Distinguish the product description from a sellable variant and its SKU. Define categories, attributes, images, availability and which values customers can filter. If fitment matters, identify the authoritative vehicle/compatibility data, its supplier and who maintains it. A make-and-model label alone does not prove compatibility.
Illustrative example, not a client result: one part may fit several vehicle configurations, while a particular finish has its own stock record. Specify the relationships and test both valid and excluded combinations. Concept Distributing's documented project provides a relevant parts-browsing example, not proof that every store needs the same architecture.
Specify price, quote and order state
Record how customer pricing is selected, when an estimate expires, who approves a quote and what becomes fixed after an order is accepted. Agree currency, tax configuration inputs, rounding and shipping eligibility with the business. Do not infer these rules from a sample product page.
Use the ecommerce payment workflow example to plan how your team would handle delayed payments and exceptions.
Define pending, paid, cancelled, refunded and fulfillment states for the actual process. Decide who may change them and what happens if the customer retries a request. A payment provider's browser return is not a substitute for an agreed server-side payment-status process.
Treat integrations as fallible systems
For shipping, payment, inventory or accounting connections, document the authoritative record, field mapping, provider access, test environment, recurring fees and failure owner. Review timeouts, retries and duplicate handling before enabling live operations.
Stripe's documentation, for example, describes duplicate and out-of-order events and idempotent API requests. A relevant integration needs its own tested handling; those provider features do not automatically make the application's entire order workflow repeat-safe.
Accept the whole workflow and own the operation
Test an unavailable variant, changed shipping destination, failed payment, duplicate event, stale stock update and staff cancellation. Reconcile order values and key reports against approved examples. Django transactions and constraints are tools for implementing rules, not substitutes for the rules themselves.
Agree who manages products, provider accounts, backups, restore tests, monitoring and updates. The ecommerce service scopes those requirements; a published package is a separate bounded offer and does not automatically include custom integrations.
Related service scope
- Ecommerce development: Review catalogue/order/integration scope.
- Business automation: Assess operational handoffs around orders.