Here is the conclusion first: what happens after transferring books between warehouses matters more than the transfer itself. The transfer order is only half the job. The other half is reconciliation — matching what left one warehouse against what arrived at the other, chasing every variance to a named owner, and closing it by a deadline. If you only confirm the transfer and walk away, you have created a window where stock exists on paper in neither warehouse, or worse, in both. My rule is simple: every transfer runs through three book states — shipped, in transit, received — and no variance sleeps overnight, crosses to another order, or gets settled by a verbal promise.

Why does this go wrong so often? Because one transfer touches two warehouses, two teams, and often two shifts. The shipping warehouse thinks, "We shipped it, so our stock should drop." The receiving warehouse thinks, "We have not finished counting it, so our stock should not rise yet." Both are reasonable, and together they create a gap: goods on a truck, recognized by nobody's books. Transfers also get split, and a receiver confirming a two-truck order at once adds cartons that have not arrived yet. I have seen this failure three times, with three different causes and the same ending: two warehouses arguing over old records at month end.

Reconciliation After a Transfer: Build the Three-State Book

The three-state book means a transfer is always in exactly one state, with no skipping and no ambiguity. State one is ship confirmation: the shipping warehouse scans pallets out, recording quantity, batch, and pallet ID, and the system drops on-hand stock while creating an equal in-transit balance. State two is in transit: that balance sits in its own account, belonging to neither warehouse, and no picking, allocation, or available-to-promise logic may touch it. State three is receipt confirmation: the receiving warehouse scans pallets one by one, adding only what was actually counted, and any gap becomes a variance record on the spot — never a "receive it all now, sort it out later."

This structure fixes ownership first. While goods sit in the shipping warehouse, that warehouse owns them. While they are on the truck, ownership follows the rule agreed in advance — usually whoever arranged the transport owns the in-transit leg, and that rule belongs in the transfer procedure, not in a post-incident argument. Once the receiver scans and signs, ownership moves — decided by scan timestamps, not by feeling.

It also moves variance discovery to receipt itself: one carton short produces one variance record carrying transfer number, pallet ID, quantity, receiver, and timestamp. Without a pallet ID, you can never trace which pallet lost the stock.

Reconciliation After a Transfer: The Step-by-Step Practice

Step one: hard-check "transferable quantity" before shipping — on-hand minus allocated, locked, and cycle-count-frozen stock, not raw on-hand. I once watched a warehouse transfer stock out of a location mid-count; the counting team found a shortage and the transfer team said the goods had left, and they spent a morning arguing. Block count-frozen and allocated stock from transfer orders in the system.

Step two: scan by pallet at shipping, never hand-key the whole order. The pallet is the smallest unit of accountability: one pallet, one code, scanned out one at a time. When people hand-key totals, they tend to enter the planned quantity as the shipped quantity. Plan 240 cartons, actually load 236, and the books say 240; the receiver counts 236, and the 4-carton gap has no clear owner. Was it short-loaded or lost in transit? Pallet scanning makes shipped quantity equal scanned quantity, so a plan-versus-actual gap surfaces at shipping, where it belongs.

Step three: publish a daily in-transit list — transfer number, days in transit, expected arrival date — with a morning reminder for anything past its expected date. Many "variances" are simply goods that arrived and sat unreceived.

Step four: receive pallet by pallet, and raise the variance at the dock. Scan the pallet code first, then count cartons and check batch, and only then confirm — a batch error is worse than a quantity error, because a quantity error can be fixed with a top-up while a batch error can mean wrong expiry or client ownership for a whole pallet. Send the variance to the shipping warehouse the same day; it has 24 hours to accept or dispute it with evidence, and silence past 24 hours counts as acceptance. Write that rule into the procedure, or records sit in "pending" forever.

Step five: close variances in three tiers, not with one blunt rule. Tier one is a small gap traceable to a specific pallet: fast top-up or book adjustment, closed the same day. Tier two is a larger gap with a known cause, such as a pallet missed at loading: a top-up linked to the original transfer, so the books show the full story. Tier three is an unknown cause: freeze the disputed quantity, forbid either side from writing it off unilaterally, and escalate to the manager with a conclusion due within three days. Delay hurts most in tier three; after a month the evidence is gone and the loss gets split.

Variance type Action Deadline Owner
Received short, pallet traceable Top-up or book adjustment Same day Decided by scan evidence
Received short, cause known Top-up linked to original transfer 3 days At-fault side
Cause unknown Freeze disputed quantity, escalate Conclusion in 3 days Manager decides
Received over Check batch, return or supplementary order Same day Receiving warehouse checks first
Batch mismatch Quarantine whole pallet, no putaway Isolated same day, resolved in 3 days Shipping warehouse checks first

Two Real Cases: How 4 Cartons Became a Much Bigger Problem

The first case comes from autumn 2024, a regional warehouse transferring seasonal goods north. The plan was 240 cartons on two trucks. The first truck's 120 cartons arrived, and the receiver, trying to save time, confirmed the full 240-carton order; the system added everything. Two days later the second truck arrived and its 120 cartons were received again, leaving the books 120 cartons over. Each side could honestly claim its numbers were correct, yet the books were wrong — only outbound records revealed the double receipt. The lesson is blunt: whole-order confirmation is an open door to double receipt, and pallet-by-pallet receipt is error-proofing, not fussiness.

The second case is more typical, and more expensive. In early 2025, a fulfillment warehouse transferred 800 cartons to a forward warehouse. Loading missed two pallets — 48 cartons — but outbound was keyed at the planned 800. The forward warehouse counted 752, a 48-carton gap, and the receiver put the stock away to "deal with it later." Later turned out to be three weeks. By then, with no variance record and no receiving evidence, neither side could prove its case, and the cartons were written off. The goods value was modest; the trust cost was larger. I have seen this pattern three times, and the script never changes: a variance not recorded at the dock becomes unprovable later.

Five Pitfalls That Wreck Transfer Reconciliation

First, mixing transfer variances with stocktake variances: one is an account between two warehouses, the other inside one warehouse, so track them by transfer number and count number separately. Second, letting in-transit stock count as available-to-promise — ring-fence it in system configuration, not by human attention. Third, variance records without time limits: without a 24-hour acceptance rule and a 3-day conclusion rule, they pile into a backlog nobody dares to touch. Fourth, reconciling quantity but not batch and expiry: right quantity, wrong batch can send the wrong goods to the right customer, and a return costs far more than a top-up. Fifth, skipping close-out, covered above — receipt is not the finish line.

The Counterintuitive Part: Faster Reconciliation Means Better Relationships

People often treat reconciliation as distrust, assuming frequent variance records will sour relations between two warehouses. The reality runs the other way: the faster a variance is raised, with complete evidence, the less the two sides second-guess each other. What damages relationships is a variance left to become a cold case. Reconciliation is not about blame; it is about turning "whose fault is this" into "whose next action is this." When every variance has an owner, an action, and a deadline, it stops being an argument and becomes collaboration.

That chain starts before shipping, of course. If you are setting up the transfer in the first place, how to avoid stockouts when transferring stock between warehouses covers safety quantities and priority sequencing; read it before planning a transfer, then use this guide for the reconciliation that closes it out.

By Lin Xiao, WMS Implementation Consultant