“The retailer does not need to build a new fulfilment network from scratch to compete.” That line, from Inside Retail's analysis of store-based delivery, is the quiet argument behind a lot of 2026 investment. As of November 13, 2026, Australia Post's research shows 59 per cent of Gen Z and Millennial shoppers will pay more for same-day or next-day delivery, while JB Hi-Fi told investors Australian sales fell 1.4 per cent from July to September and slowed again in October. Demand for speed is rising in a softer market, which leaves the store network as the cheapest lever available.
The depot is where the promise is won or lost
Most retailers can quote their national availability number. Fewer can say how long it takes a product to travel from the backroom to the shelf in a specific store. That second number is the one that decides whether a delivery promise placed at 9am can be kept at 2pm, and it is the number best-in-class operators have been cutting: 23 per cent against 4 per cent for everyone else, according to the Zebra and Aberdeen research.
The mechanism is unglamorous. When store, warehouse and delivery teams work from separate records, an associate cannot see that stock exists two aisles away in the backroom, so the shelf stays empty while the system reports positive inventory. Connecting those records does not add capacity, but it converts stock that was already paid for into sellable units.
Speed is a routing decision, not a fleet decision
| Lever | What changes | Cost profile | Watch metric |
|---|---|---|---|
| Ship from store | Orders served from local stock | Low, uses existing space | Cancellation rate |
| Click and collect | Customer sets the window | Very low | Collection conversion |
| Locker handoff | Delivery decoupled from presence | Low | Failed delivery rate |
| Dedicated fleet | Capacity bought outright | High, fixed | Cost per order |
The expensive lever is usually the last one retailers should pull. A dedicated fleet buys capacity whether or not orders arrive, which is a poor trade in a quarter when discretionary spending is softening. Shipping from stores and offering collection windows uses capacity that already exists, and it moves volume away from the most expensive delivery slots without telling the customer no.
Collection options also absorb the timing problem. A shopper who chooses a locker or a store pickup is not waiting at home, so a two-hour variance stops being a failure. That is why fulfilment choice tends to outperform pure speed on satisfaction: it gives the customer control rather than a promise the retailer has to hit to the minute.
Labelling is a cost line disguised as admin
A mislabelled parcel cannot be read by the sorter, so it circulates and occupies a cell another parcel could have used. During peak, that single error compounds into a delay that looks like a network problem and is actually a process problem. Checking labels and packaging before volumes build is cheaper than adding a shift in December.
Availability belongs in the trade area, not the annual report
National availability hides the only figure store teams can act on: availability inside the catchment a given store serves. Two stores with the same national score can behave completely differently if one serves dense residential streets and the other an office district with a lunchtime spike. Reporting by cluster makes replenishment and range decisions possible; reporting by country makes them invisible.
A routine that fits inside a week
Start with a weekly cluster report: availability, replenishment cycle time and cancellations, side by side. Then separate true stockouts from listing gaps, because the fixes differ — one needs an order, the other needs a batch update. Then cap delivery promises at store-level inventory rather than warehouse inventory, so the checkout page never sells what the shelf cannot support. Finally, review labelling exceptions alongside the delivery numbers, since they explain a share of late parcels that otherwise get blamed on carriers.
None of this requires new buildings. It requires one version of the truth about stock, shared by the people who pick, the people who deliver and the page that takes the order. In a market where shoppers will pay for speed but are watching discretionary spend, that is the difference between a promise and a cost.
FAQ
Is ship-from-store cheaper than a fleet?
Usually, because it uses space and stock that already exist.
Do shoppers accept slower promises?
Yes, when they choose the window themselves.
Where do late parcels start?
Often at labelling and packaging, before the carrier touches them.
What should be reported weekly?
Availability, cycle time and cancellations by cluster.
