Stockout or Air-Freight Expedite? A Break-Even Decision Model

The short answer: price only the shortage the air bridge can prevent

Do not compare an air-freight quote with the total revenue forecast for the stockout period. Compare a documented no-expedite baseline with one or more feasible air-bridge lots. Approve a lot only when the contribution and other evidenced costs it is expected to preserve exceed its incremental cost under the business's stated decision policy.

Three boundaries keep the calculation honest:

• Count only demand that would otherwise go unfilled after the air stock becomes available to the affected Australian location or channel and before normal replenishment becomes available.
• Distinguish sales likely to be permanently lost from orders that will be backordered, preordered, deferred or substituted.
• Use a current all-in quote and subtract only baseline costs that will definitely disappear if the bridge proceeds.

This is an exception decision. The broader air-versus-sea freight guide owns the general mode comparison. The reorder-point and safety-stock guide owns the proactive replenishment trigger. Here, a normal plan already exists and a specific SKU is projected to run out before replacement stock becomes usable.

Freeze the no-expedite baseline

Write down the counterfactual before negotiating the rescue so later changes in demand or inbound timing remain visible.

Define one SKU, fulfilment location and selling channel. Record usable on-hand inventory, commitments already promised, the projected zero-stock date, the normal inbound quantity, and base and downside dates for that inventory to become available for sale or use. Timestamp the snapshot and name the owner of each input.

Stock elsewhere is not necessarily available to the affected channel. Quarantined, damaged, allocated or unreleased stock and open purchase orders are not usable inventory.

Separate the expected out-of-stock response:

Oracle's retail inventory documentation distinguishes a backorder—inventory already sold but not fulfilled—from lost sales. Shopify's out-of-stock documentation also shows that a merchant can block purchase at zero stock or continue selling. Those are useful distinctions, not proof of a particular business's cancellation or lost-sale rate.

Use observed order and customer behaviour where it exists. Otherwise show a range; do not fill the gap with generic rank, churn or ecommerce benchmarks.

Quote feasible bridge lots, not an arbitrary unit total

Test whole cartons or pallets that the supplier can separate without delaying the normal shipment. Respect the actual pack quantity.

Quote more than one feasible lot when fixed charges or price breaks could change the answer. Record:

• cargo-ready date, origin pickup point and Australian delivery point;
• final piece count, dimensions, gross weight and stackability;
• chargeable weight and the provider's calculation and rounding rule;
• service, routing, validity, acceptance and cut-off assumptions;
• included, estimated, pass-through and excluded charges; and
• base and downside dates through customs, biosecurity, delivery, receiving and channel availability.

Qantas Freight explains that its basic air charges use the higher of actual and cubic weight and publishes a carrier-specific international conversion. The provider's rule controls. Use the air-freight chargeable-weight guide to verify the arithmetic; do not apply Qantas's factor to another service.

Calculate incremental expedite cost

Start with the current all-in bridge quote. Add incremental split, repacking, documentation, handling, change, cancellation and receiving costs. Then subtract costs from the no-expedite path only when they will definitely no longer be paid. Reconcile every cost line to one side of the comparison once: bridge cost, added change cost, definitely avoided baseline cost or separate one-off benefit.

IATA's air-cargo tariff overview notes that carrier tariffs can exclude pickup, delivery, storage, insurance, customs, duties, taxes and other services unless stated. A base rate per kilogram is therefore not the decision cost.

If the original sea container still costs the same after 12 cartons are removed, there is no avoided sea-freight line for those cartons. If origin or destination minimums apply twice, include both. Treat duty, GST, insurance, clearance and financing as differences only when the two paths genuinely change them and the treatment has been confirmed. Do not subtract a cost here and also add the same amount as an avoided benefit later.

The working equation is: incremental expedite cost = all-in air bridge + added change and handling costs - baseline costs definitely avoided.

Cap the benefit at preventable shortage units

Air cannot recover demand lost before the bridge becomes usable. It also does not earn a stockout-avoidance benefit for inventory left after the normal inbound becomes usable.

For each scenario, calculate the bridge window from the air available-to-sell date up to, but not including, the normal replenishment available-to-sell date. Estimate only demand that would otherwise remain unfilled inside that window. Then cap the result at the accepted bridge units available to the affected channel. Use the same calendar basis for dates and demand so a daily forecast does not acquire an extra day at either boundary.

The equation is: preventable shortage units = the lower of accepted bridge units and otherwise-unfilled demand inside the bridge window.

If 300 units could go unfilled but only 180 accepted units reach the channel, the cap is 180. If only 90 would otherwise go unfilled, the other 90 receive no stockout-avoidance value without separate evidence.

Use base and downside dates rather than one marketed transit number. Supplier release, a missed connection, an inspection or a receiving backlog can shorten the bridge window.

Separate lost, backordered and deferred demand

Zero stock can produce a lost sale, preorder, backorder, substitution or abandonment. These outcomes have different values.

Count direct retained contribution only for sales expected to disappear without the bridge. Backordered or deferred sales do not lose their full contribution if they are still expected to complete. Define the permanently-lost share against the otherwise-unfilled demand in the tested channel and window, using a comparable observed cohort where possible. Do not transplant a rate from a channel that blocks checkout into one that accepts preorders or backorders. Add an evidenced discount, service credit, cancellation cost or extra handling cost separately when it genuinely changes between the two paths.

Define contribution per retained sale consistently with finance as selling price less the variable sale and fulfilment costs that would not be incurred if that sale disappeared. Do not use gross revenue. Keep the air premium in incremental expedite cost rather than subtracting it again inside contribution, and keep a discount, credit or handling consequence out of contribution if it is recorded separately.

A historical US Federal Highway Administration methodology distinguishes lost-sale and backorder costs and says lost-sales cost varies by business context. It provides no current Australian percentage.

Customer lifetime value, churn, marketplace rank, advertising recovery and reputational harm default to zero. Include them only with proprietary evidence, an approved method and a double-counting check.

Calculate the break-even case

For each candidate lot and scenario:

• Expected retained units = preventable shortage units × evidenced permanently-lost share.
• Expected retained contribution = expected retained units × contribution per retained sale.
• Expected preventable value = retained contribution + documented non-duplicative one-off costs avoided.
• Net expedite value = expected preventable value - incremental expedite cost.

When quote cost is fixed for the tested lot, the break-even expected-retained-sales threshold is (incremental expedite cost - other documented costs avoided) ÷ contribution per retained sale. Divide the unrounded result by the evidenced permanently-lost share, then round the resulting preventable-demand threshold up to a whole unit. If the business instead requires a conservative whole-retained-sale hurdle, state that separate policy and its rounding order. Requote when quantity, packout, chargeable weight, route, service, cargo-ready date or validity changes.

Do not convert this scenario threshold into a permanent SKU rule. The next exception may have a different channel, customer response, quote, pack density or recovery date.

Fictional example: a 240-unit bridge

The following example is fictional. Its dates, quantities, demand, lost-sale share, margin and costs are not OPL benchmarks, market rates or promises.

An Australian online seller projects that SKU-A will reach zero available inventory on 5 October 2026. Normal replenishment is expected to become available on 19 October. The supplier can separate 12 cartons containing 20 units each without delaying the balance.

Base case

The air stock becomes available on 8 October. Counting 8 October through 18 October, before normal stock is usable on 19 October, gives an 11-day bridge window. Forecast demand inside it is 286 units. The bridge quantity caps preventable shortage units at 240.

At the fictional 60% permanently-lost share, expected retained units are 144. At A$22 contribution each, retained contribution is A$3,168. Assume a documented A$500 customer service credit is also avoided only if stock is available by the base-case date. That credit is not included in the A$22 contribution or the A$1,750 baseline-cost deduction. Expected preventable value is A$3,668, producing a net expedite value of A$218 after the A$3,450 incremental cost.

The expected-value break-even is (A$3,450 - A$500) ÷ (A$22 × 60%) = 223.48, rounded up to 224 preventable units. Rounding the intermediate 134.09 expected retained sales to 135 first would create a separate, conservative 225-unit policy threshold. The 240-unit base case clears both.

Downside case

The air stock becomes available on 12 October. Counting 12 October through 18 October gives a seven-day bridge window, exposing 182 units of demand before normal stock becomes available. That is lower than the 240-unit lot, so only 182 units receive stockout-avoidance value.

Expected retained units are 109.2, a probability-weighted value rather than a fraction of a physical sale. Contribution is A$2,402.40. The credit deadline is missed, so net expedite value is negative A$1,047.60. Break-even is 262 preventable units, beyond both the 182-unit window and 240-unit lot.

The decision flips. The correct disposition is not automatic approval: it is conditional hold. Requote a smaller whole-carton lot, improve the end-to-end date evidence, or have the named decision owner accept the downside under the business's policy.

Check the Australian available-to-sell path

Air transport does not remove Australian import controls. The Australian Border Force import guidance explains that declarations may be required to clear sea or air cargo from customs control. The Department of Agriculture, Fisheries and Forestry states that imported goods may be released or directed to inspection, treatment, isolation or a hold pending more information.

The calculation therefore needs a usable-date range that includes documents, customs and biosecurity processes where applicable, terminal or depot release, final delivery, receiving and channel availability. It must not substitute airport arrival for saleable receipt.

Shopify's purchase-order transfer documentation separates the purchase order from movement and receiving: accepted units become available at destination; rejected units do not. An importer's systems may differ, but the control remains the channel's genuine fulfilment date.

Confirm cargo acceptance before approving the economics. Special cargo can need shipment-specific documentation, packing or carrier review. Hold unresolved acceptance, customs, biosecurity or product-compliance matters.

Record approve, reject or conditional hold

The approval record should be short enough to use during an exception and complete enough to audit later.

Record the SKU, location, carton count, quote validity, chargeable basis, usable-date cases, forecast, response evidence, contribution definition, result, owner and reopen triggers.

Fix the planning cause after the exception

An approved expedite can be commercially correct and still reveal a planning failure. After the stock is received, replace estimates with actual dates, costs, accepted quantities, demand outcomes, backorders, cancellations, credits and contribution preserved. Compare the result with the approval case.

Then return the learning to the planning controls. Review the reorder point and safety stock, the demand forecast, supplier readiness and production milestones. Do not hide a recurring lead-time or forecast problem inside a permanent expedite budget.

The decision is successful when it makes one exception explicit: what shortage could be prevented, what it was worth, what the bridge cost, what uncertainty remained and who accepted it. It is not successful merely because an aircraft moved the cartons.

Sources

• Qantas Freight: Calculate freight costs
• IATA: Air cargo tariffs and rules
• Oracle Retail: Inventory Planning glossary
• Shopify: Selling out-of-stock products
• Shopify: Creating an inventory transfer for a purchase order
• Australian Border Force: How to import
• Department of Agriculture, Fisheries and Forestry: Clearance and inspection of goods
• US Federal Highway Administration: Freight Benefit/Cost Study, Phase III

Sources retrieved or rechecked 1 September 2026 Australia/Sydney. This article provides a general operational decision model, not shipment-specific freight, customs, biosecurity, dangerous-goods, legal, tax, accounting or financial advice. Verify current quotes, acceptance, import requirements and available-to-sell dates for the actual shipment.