Backorder rate is the share of orders a storefront accepts that cannot be fulfilled immediately from on-hand inventory and are instead promised to ship once stock arrives.
Unit-weighted form (backordered units / units ordered) used when a single large pre-order would distort the order-count view.
Some operators run a unit-weighted version (backordered units divided by units ordered) when a single large pre-order would distort the order-count view; the order-weighted version is the more common headline.
The load-bearing distinction is order-state versus SKU-state. A stockout is a SKU-state condition: the SKU was unavailable to add to cart for some window. A backorder is an order-state condition: the SKU was sold while flagged as available, with a delayed-ship promise attached to the order. The two metrics are siblings and can move independently. A brand that aggressively flips SKUs to unavailable the moment on-hand hits zero will run a high stockout rate and a near-zero backorder rate. A brand running a pre-order program against incoming inventory can post a 0% stockout rate and a non-trivial backorder rate at the same time, because the storefront never showed the SKU as unavailable.
Four operational drivers push the rate up. The intentional one is a pre-order program: the brand accepts orders against a PO that has not yet landed, trading immediate fulfillment for early demand capture. The accidental one is overselling caused by a stale inventory feed between the warehouse system and the storefront — orders flow in against units that physically left the building hours ago. Delayed inbound shipments from suppliers extend the backorder tail on units the brand expected to ship from today’s PO. Demand spikes that outrun safety stock are the fourth: a campaign hits, sell-through outpaces the coverage buffer, and the next wave of orders ships against the next inbound.
The customer-experience tail is the cost. Backorders that ship late drive WISMO (where-is-my-order) contact volume to support, and operators commonly see cancellation and refund rates climb as the gap between order and ship extends. The operator lever sits at the storefront-inventory sync: accept orders to hold demand and accept the cancellation tail, or flip the SKU to unavailable when on-hand hits zero and lose the demand capture. Neither posture is universally right — the call depends on margin, customer expectation, and confidence in the inbound date. Backorder rate and fill rate sit on opposite sides of the same promise: fill rate counts the orders that shipped from stock; backorder rate counts the orders that didn’t but were kept on the books.