Inventory by location, not a global product count

Alternatives considered

Discarded

Keep a global quantity per product.

Chosen

Track stock by location and preserve the movement history.

Decision made

Model inventory transactionally by location and product, with a full movement history, instead of a single global scalar quantity.

Rationale

The decision reflects the planned bazaar and wholesale operation across multiple locations, and avoids losing the history needed for later auditing.

Frequently asked questions

What did Pymerce decide about "inventory by location, not a global product count"?

Model inventory transactionally by location and product, with a full movement history, instead of a single global scalar quantity.

What kind of inventory system makes sense if I have multiple warehouses or branches?

One that tracks stock by location instead of a single number per product. Pymerce models inventory transactionally by location and product, with a full history of movements, so a sale at one branch never deducts stock that's physically sitting at another.

What are the main approaches to tracking stock, and which one does Pymerce use?

Broadly speaking, there are two: a global quantity (one number per product) or a transactional record by location with a movement history. Pymerce uses the second approach, because it needs traceability for auditing later on, and because the operation it's built for — bazaar and wholesale — already spans multiple locations from day one.

Why isn't a global quantity per product enough?

A global quantity doesn't tell you which location the stock is physically in, and it doesn't keep a history of movements. That leads to overselling between branches and makes it hard to audit what happened to inventory — something Pymerce needed to solve at the design stage, not patch later.