Case Study · Retail
Optail Inter-Store Transfer: size-run consolidation, out-of-collection and season-end transfers
Optail's first module. It started with block transfer in February 2024; on top came the consolidation transfer that gathers broken size runs, the out-of-collection transfer that moves product to stores that do not carry it, closing-store clear-out, and the class-based transfer that concentrates product in selling stores at season end. Transferred goods sell through within 21 days at five times the manual rate.
- ~25%
- Sell-through
- 5
- Transfer types
- ~65%
- Automation
- 12.5%
- Capacity ceiling
In 21 days, summer 2025; manual ~5%, last year ~12%
On the same scenario infrastructure
Summer 2025 season
Max a receiving store can take, relative to its stock
Problem
A product waiting on the shelf in one store is being asked for and not found in another; size 38 of the same model has sold out in one store while three pairs sit in another. Planning transfers by hand is impossible for hundreds of stores and thousands of options; when it is done it takes weeks, and whether the goods sent actually sold where they went is unknown.
Transfer became Optail’s first module — because it was the most repeated and least measured decision.
One infrastructure, five transfer types
The user sets up the scenario: gender, main category, season group; sending (A) and receiving (B) stores or store groups; receiver and sender cover; exceptions. The scenario is saved under a unique id; its input, parameters and output can be traced years later. Five transfer types share the same flow with different rules:
| Type | What it does |
|---|---|
| Block transfer | Moves an option as a whole from a low-performing store to a high-performing one |
| Size-run consolidation | Gathers broken options — few sizes, few units left; receivers chosen by sell-through, Pareto and sales priority |
| Out-of-collection (NoC) | Moves an option to a store that does not carry it but sells fast in that class; widens the collection |
| Store clear-out | Distributes a closing or converting store’s stock to the most suitable receivers |
| Season end | Size-run-based gathering + class-based widening from September 2026; the warehouse is not involved, goods flow from A stores to B stores |
A block transfer can be converted into consolidation or block+consolidation; it cannot become a clear-out — which type can turn into which is defined in a table, so no scenario junkyard forms.
The algorithm, step by step
The flow, following the season-end procedure:
- Universe. The option list is drawn from the scenario’s store and product combinations; only SKUs that were actually shipped before remain.
- Elimination. Window products, options with total stock below the threshold, products on the exclusion list, receiver/sender group rules from the exception engine; a store that sent this option through Optail in the last 21 days cannot receive the same option. Every eliminated option is written to the report with a reason code.
- Tagging. Sender candidate: sizes in stock and total units below the threshold, nothing in transit. Receiver candidate: a store that carries the option in its collection.
- Need. Target stock per receiving SKU — core sizes 2 in footwear, the others 1; need = target − (on hand + in transit).
- Priority. Receivers are ranked by 56-day sales, season sales, selling above average while already carrying the option, and need.
- Loop. Options in order of sales; for each receiver, sizes by need; among senders, the store with the most sizes goes first. When a receiver’s capacity counter — 12.5% of its total stock — is full, that store stops receiving.
Shelf life is 21 days: newly arrived product is not sent back. The class-based block runs after the first block ends and sees its results — the first completes the existing collection, the second widens it, provided an assortment can still be built from what is left in the A stores.
The approved proposal goes to AX as a transfer order; the sending store picks, the receiver accepts.
Measuring
Every transfer’s 21-day sell-through is tracked: how much of what was sent sold within three weeks. In the summer 2025 season roughly a quarter of Optail transfers sold through in 21 days; for manual transfers the rate was around 5%, and 12% in the same period the year before. About two thirds of transfers came from Optail. The stores that receive and send the most are reported every season; a repeatedly sending store is a signal for the initial allocation plan.
Why this way
- Decision at option, quantity at SKU. The transfer decision is made at product+colour level and sizes are distributed by need; doing it the other way round moves the break, it does not fix it.
- The eliminated are reported too. Which option did not enter the transfer and why — the planner argues about the reason, not the result.
- Types convert, they do not mix. Five types on one infrastructure; but which scenario can become which is defined.