Binational Inventory: What One Dashboard Across Two Warehouses Has to Reconcile

Holding stock on both sides of the border sounds like a redundancy decision and it is not. It is a decision about which units have already been imported and which have not, and those 2 populations behave differently in almost every calculation a seller runs.

This article covers what a single view across a US warehouse and a Mexican warehouse actually has to reconcile, how the reorder point changes when the lead time contains a customs step, and where the split stops paying.

The split is a tax decision before it is a location decision

"El impuesto se calculará aplicando a los valores que señala esta Ley, la tasa del 16%."

Ley del Impuesto al Valor Agregado, Article 1

Every unit that crosses becomes an import, and the import is where the duty and the 16 percent IVA are settled. A unit sitting in Laredo has not triggered either. A unit sitting in Monterrey has triggered both, and the money is already out of the business.

That is the whole difference in 1 sentence. Stock on the US side is cheaper to hold and slower to deliver. Stock on the Mexican side is faster to deliver and has already consumed cash. A binational position is a way of choosing where on that line each product should sit, product by product, rather than choosing once for the catalog.

Units on the 2 sides are not interchangeable

A dashboard that shows a single total is worse than no dashboard, because it invites a promise the operation cannot keep. 400 units of an item, 300 in Laredo and 100 in Monterrey, is not 400 units of same week delivery. It is 100 units of same week delivery and 300 units that are 1 customs entry away.

The reconciliation has to hold both facts at once, which means the record for each SKU carries a location dimension and a customs status dimension rather than a quantity.

For sellers on Amazon specifically, the placement decision is worked through in Amazon Mexico Inventory Placement: NARF Versus Local FBA, Cost Per Unit.

What 1 dashboard has to reconcile

The bottom 2 rows are the ones that get discovered late. Moving stock north is not the reverse of moving it south. It is a separate export with its own paperwork, and the duty already paid does not come back by default.

The reorder point changes when the lead time contains a customs step

"En México solo un agente aduanal con patente puede presentar el pedimento."

Servicio de Administración Tributaria

A domestic reorder point is built on a lead time with a narrow distribution. A cross border one is not. The driving portion between Laredo and Monterrey is roughly 3 hours, a shipment cleared on the green light lands in 1 to 2 business days, and a physical inspection on the red light adds 1 to 2 days.

So the transport time is not the problem. The customs time is bimodal, and a reorder point set on the average of a bimodal distribution is wrong in both directions. It is too high on the majority of transfers and too low on the ones that get inspected.

The workable approach is to set the safety stock against the slow branch rather than the average, and then to reduce the number of branches taken by transferring less often in larger batches. That is the opposite of the just in time instinct, and on this lane it is correct.

Split the catalog, not the quantity

The common mistake is to hold a percentage of every SKU on each side. It produces the worst of both: cash committed on slow movers and not enough depth on fast movers.

A better split runs on 3 rules.

• Fast movers with settled classification go south. They turn quickly enough that the committed cash comes back, and their tariff position is not in question.
• Slow movers and long tail stay north. An item selling 4 units a month does not justify importing 40 of them to save 2 days.
• Anything with an unsettled classification stays north until it is settled. Mexico's tariff line runs to 10 digits, and importing 500 units under a code you are not sure of concentrates a correction into 1 large entry.

The split is then reviewed on velocity rather than on a calendar, because the point of the position is to follow demand rather than to describe it.

Reviewing it on velocity also catches the seasonal case. An item that is a slow mover for 9 months and a fast mover for 6 weeks should not live on the same side of the border all year, and a calendar review set for January will decide that question at exactly the wrong moment.

Amazon storage cost is seasonal, which moves slow stock

For sellers using local fulfillment inside Mexico, the storage cost is not flat across the year. Monthly storage runs 0.36 MXN per cubic decimeter from January to September and 0.53 MXN from October to December, and the long term surcharge starts on day 181.

Two consequences follow. Slow moving stock imported in the first half of the year is paying the higher rate exactly when it is least likely to sell out of it, and an item that has not moved by day 181 has crossed into a penalty that will not reverse on its own.

A binational position is the natural answer to both. Depth for the peak sits north until it is needed, and the units that would have aged past 181 days never take up the expensive space in the first place.

Fixed customs charges reward fewer, larger transfers

Two charges do not scale with the size of the shipment. The customs processing fee is 8 per thousand of customs value with a floor of 258.91 pesos, and that floor applies below a customs value of roughly 32,364 pesos. Brokerage runs 3,500 to 15,000 MXN per entry.

A replenishment run as 4 small transfers pays those charges 4 times, roughly 15,036 pesos at the minimum. The same volume moved once pays them once, roughly 3,759 pesos. Across 12 months that is the difference between a fixed cost that disappears into the unit economics and 1 that shows up in the margin.

This is the strongest argument for holding the buffer north rather than transferring frequently. The buffer costs storage. The frequency costs entries, and entries have a floor.

Returns get cheaper on the southern side, and that is often the whole case

"Para los efectos del párrafo anterior la garantía no podrá ser inferior a noventa días contados a partir de la entrega del bien o la prestación del servicio."

Ley Federal de Protección al Consumidor, Article 77

The statutory warranty runs from delivery and cannot be shorter than 90 days, which means a seller shipping into Mexico is carrying a return obligation for at least 3 months on every order. Where that return lands is decided by where the stock was held.

An order fulfilled from a Mexican warehouse comes back to a Mexican address. It is a domestic movement, the buyer pays nothing to send it, and the unit can be inspected and put back on the shelf in the same building it left. An order fulfilled across the border and returned to the US is an export from Mexico, and it needs an entry to get back into US stock.

That asymmetry rarely appears in the spreadsheet that decides the split, and on a category with a return rate above 10 percent it can be larger than every other line in the comparison. A unit returned domestically is recoverable inventory. A unit returned across the border is frequently written off, not because it is damaged but because the cost of retrieving it exceeds what it is worth.

So the honest version of the split test has 3 inputs rather than 2: how fast the SKU turns, how much cash the import commits, and how often it comes back. The third input is the one that most often flips the answer.

The 4 numbers to review monthly

• Committed cash on the Mexican side. Goods cost plus duty plus IVA plus entry charges, on units that have not sold. This is the real cost of the position.
• Days of cover on each side, separately. A blended figure hides a stockout that is 1 customs entry away from being solved.
• Entries per month. If this number is climbing while volume is flat, the transfer batch is too small.
• Units aged past 181 days on the Mexican side. The surcharge threshold, and a direct signal that a SKU was placed on the wrong side.

The storefront side of the same split, meaning what has to be configured before you divide anything, is covered in Shopify Into Mexico: Splitting Stock Across Two Warehouses Without Running Out.

How BringGo Ship Handles This

BringGo Ship holds stock for sellers at our Laredo facility and inside Mexico, so the transfer between the 2 is a routine internal movement rather than a new provider relationship. Our team files the entry, coordinates with a patented customs broker, and reports the units by location and by customs status rather than as a single total, which is what lets a seller promise a delivery date they can actually hold. Transfers are batched deliberately, because the fixed charges per entry are the part of this that rewards patience.

Frequently asked questions

Should I just hold everything in Mexico? Only if the catalog turns fast enough to justify the committed cash. Duty and 16 percent IVA are paid on arrival rather than on sale, so a slow SKU held south is a loan to the tax authority.

Can I move stock back to the US if I get the split wrong? Yes, and it is an export from Mexico with its own entry. Treat it as a correction with a real cost rather than a reversible setting.

How often should I transfer? Rarely enough that the fixed charges per entry stay small against the batch value, and often enough that the southern buffer does not run dry during the customs slow branch. In practice that lands on a monthly rhythm for most sellers rather than a weekly one.

Does 1 dashboard mean 1 provider? Not necessarily, but 2 providers means somebody has to reconcile 2 records, and in practice that somebody is the seller. The value of a single view is that the reconciliation has already happened.