Contents
- Why does one inventory number stop working at the border?
- Which three inventory states does nobody configure?
- Which channel should read which location?
- Setting a buffer that matches your transit, not your instinct
- What should you reconcile, and how often?
- Where cross-border oversells actually come from
- How we keep both counts honest
- Definitions
- Frequently asked questions
- Should I use one Shopify location for Mexico or one per warehouse?
- Can I just sell the total and sort it out when orders come in?
- How large should the safety buffer be?
- How do I count units that are sitting at the border?
- How often should I reconcile?
- Related Reading
- Sources
Give each warehouse its own Shopify location and stop treating the total as available. A unit in Texas cannot fill a Mexico City order tomorrow. Point the Mexico channel at the Mexico location only, size the buffer against your measured transit, and reconcile both counts on a fixed schedule.
- Overselling across a border is rarely a software failure. It is usually one location count standing in for two physically separate pools of stock.
- Units that have left the US warehouse but have not been received in Mexico belong to neither count, and that gap is where most oversells are created.
- A cross-border transfer takes days, so the buffer that protects a domestic seller for a few hours has to protect you for the length of the whole transit.
- The oversell shows up as a cancellation, and the cancellation is what damages the selling account, not the inventory error itself.
Why does one inventory number stop working at the border?
Because availability is not a quantity, it is a quantity plus a delivery promise, and the promise changes by location.
A single-warehouse seller can treat inventory as one number because every unit carries the same promise. Ship from the one warehouse, arrive in the same window, done. Add a second warehouse on the other side of an international border and that stops being true. Forty units in Laredo and forty units in Monterrey are not eighty available units for a customer in Guadalajara. They are forty units that can be delivered this week and forty units that require an export, a border crossing, an import clearance and a receiving step before they can be delivered at all. Shopify will happily show eighty. It has no opinion about customs. If your Mexico storefront sells sixty on a good day, twenty of those orders are promises you cannot keep on the timeline the customer was shown. The failure is not visible on the day it happens. It becomes visible days later, as a late shipment or a cancellation, and by then the cause looks like a logistics problem rather than a configuration one.
Which three inventory states does nobody configure?
On hand, in transit and received. Most setups only model the first and the last.
Walk any unit through a cross-border move and it passes through three states. On hand in the origin warehouse. Physically present, countable, sellable to any market the origin warehouse can serve. In transit. It has been picked and loaded and it has left. It is no longer in the origin count. It is not yet in the destination count. On most setups it is invisible, and invisible inventory is the single largest source of cross-border oversells, because the same units get sold twice: once as origin stock before dispatch, once as destination stock before arrival. Received in the destination warehouse. Checked in, counted, sellable locally. The fix is not clever software. It is refusing to let transit be invisible. Some operators create a third Shopify location named for the transfer itself and move units into it on dispatch and out of it on receipt. Others simply subtract in-transit quantity from the origin location the moment the truck leaves and add it to the destination only on confirmed receipt. Both work. What does not work is leaving a multi-day gap unmodeled and hoping the sales rate stays low.
Which channel should read which location?
The Mexico channel reads the Mexico location only. Anything else is a promise you cannot keep.
This is the setting most sellers get wrong, and it is a single decision rather than a project. If you sell into Mexico through a Mexico-facing storefront or marketplace listing, that channel should be pointed at the Mexico location and nothing else. The Texas stock is not a reserve you can dip into for a same-week Mexican order. It is next month's Mexican stock that happens to be sitting in Texas today. The objection is always the same: pointing the channel at one location means showing a lower available number, and a lower number means fewer sales. That is true and it is the right trade. An order you cannot fill on time is worth less than no order, because the unfilled one costs you a cancellation, a refund, a possible negative review and a metric hit, while the order you never took costs you nothing but the margin you were never going to keep. If the Mexico number is too low to sell against, that is a replenishment problem, and it should be solved by moving stock earlier rather than by promising stock that is in another country.
Setting a buffer that matches your transit, not your instinct
The buffer has to cover the time between counting and delivering, which across a border is days, not hours.
Every operator eventually sets a safety buffer, a quantity held back from the sellable number so that a small counting error does not turn into an oversell. The common mistake is to carry over a domestic instinct. A domestic seller might hold back a couple of units because the warehouse count refreshes a few times a day and the worst case is a few hours of drift. Across a border the drift window is the entire transfer, and a transfer includes export paperwork, a crossing, an import clearance and a receiving step, any of which can add a day on its own. So size the buffer against the actual measured window between your last reliable count and the moment a unit becomes sellable in the destination. Measure that window on your own shipments rather than assuming it. If it is four days and you sell ten units a day in Mexico, a two-unit buffer is decoration. Write the buffer down as a number with a reason attached, review it when your transit time changes, and treat any oversell as evidence that the buffer was wrong rather than that the warehouse miscounted.
What should you reconcile, and how often?
Compare the platform count to the physical count on a fixed schedule, per location, and investigate every difference.
Reconciliation is the step that turns a configuration into an operation, and it is usually the step that gets skipped. Pick a fixed cadence. Weekly is enough for most sellers moving a few hundred units a month; daily is warranted during a peak. On that cadence, pull the platform quantity per location and the warehouse quantity per location and put them side by side. Then do the part that matters: investigate every difference, including the small ones. A one-unit gap is not noise. It is either a miscount, an unrecorded return, a damaged unit that was never written off, or a transfer that was recorded on one side and not the other, and each of those has a different fix. Sellers who write off small gaps as rounding discover at the end of a quarter that the gaps were all the same cause and it was fixable in week one. Keep the record. A reconciliation history is what lets you tell a systematic problem from a bad day.
Where cross-border oversells actually come from
| Cause | How it looks when it fails | The fix |
| Both warehouses counted as one pool | Order placed against stock in the wrong country | Separate Shopify location per warehouse |
| In-transit units counted in both places | Same unit sold twice, days apart | Model transit as its own state |
| Mexico channel reading total inventory | On-time promise made against US stock | Point the channel at the Mexico location only |
| Buffer sized for domestic drift | Small counting errors become cancellations | Size the buffer against measured transit |
| No scheduled reconciliation | Error found by a customer, not by you | Fixed cadence, per location, investigate every gap |
How we keep both counts honest
BringGo Ship operates warehouses on both sides of this route, in Laredo, Texas and in Monterrey, and we treat the transfer between them as a state we report rather than a gap you have to guess at. When your stock leaves our US facility we tell you it has left, and when it is received and counted in Mexico we tell you that too, with the quantity we actually put on the shelf rather than the quantity on the packing list. That difference is the number most sellers never see, and it is the one that decides whether your Shopify count is trustworthy. If you want to compare your platform quantity against our warehouse quantity on a fixed schedule, we will send it on that schedule.
Definitions
- Location: A place Shopify holds a separate inventory count for, which should map to one physical warehouse and not to a country or a region.
- In transit: Stock that has left the origin warehouse and has not yet been received at the destination, and therefore belongs to neither location's sellable count.
- Safety buffer: A quantity deliberately withheld from the sellable number so that ordinary counting drift does not produce an order you cannot fill.
- Reconciliation: The scheduled comparison of platform quantity against physical warehouse quantity, per location, with every difference investigated.
Frequently asked questions
Should I use one Shopify location for Mexico or one per warehouse?
One per physical warehouse. A location is meant to answer the question of where a unit is and how fast it can reach a customer, and a country is not specific enough to answer either. If you hold stock in Laredo and in Monterrey, that is two locations even though one of them serves the Mexican market indirectly.
Can I just sell the total and sort it out when orders come in?
You can, and it works until your Mexican sales rate exceeds what the Mexican warehouse holds. At that point every extra order is a late shipment or a cancellation, and cancellations damage the selling account in a way that the missed sale never would.
How large should the safety buffer be?
Large enough to cover your measured transit window at your current sales rate. Multiply your daily units sold in Mexico by the number of days between your last reliable count and the moment stock becomes sellable there. Anything smaller is a guess.
How do I count units that are sitting at the border?
As in transit, which means not sellable in either location. Units in clearance are the clearest case of inventory that exists but cannot be promised, and treating them as available is the most common single cause of a cross-border oversell.
How often should I reconcile?
Weekly under normal conditions and daily during a peak. What matters more than the frequency is that every difference gets investigated rather than written off, because small gaps usually share one cause.
Related Reading
- what a Mexican warehouse checks on receipt
- how dwell time at Laredo affects your transit window
- choosing between LTL and FTL for replenishment
Get our warehouse counts on a schedule you set
Sources
- Shopify Help Center, Managing inventory locations (help.shopify.com)
- ANAM, Mexican National Customs Agency (anam.gob.mx)
Note: This content is for general information only and is not legal, tax or customs advice. Rates and rules can change often in 2026; verify the current details with an official source (SAT, DOF, CBP) or a licensed customs broker before acting.
How this was made: The BringGo Ship team chose the topic and the sources. The text was drafted with AI assistance, its figures and regulatory details were checked against official sources (DOF, SAT, ANAM, CBP), and the article was reviewed by our team before publication.
James Carter
Warehousing and Fulfillment Operations
Writes on Amazon Mexico and e-commerce fulfillment across the Laredo border.
View profile