Cross-docking e-commerce: sell stock you don't hold

Logistics has known cross-docking for a long time. Maersk defines it as "a process where products from suppliers are directly transferred to a customer, with minimal to no storage time in between": storage costs fall sharply, and so do handling costs. What that definition leaves out is how different the e-commerce version is from a cross-dock dock in mass retail. A distributor receives pallets planned far in advance. An e-merchant triggers a supplier purchase on every customer order, for a product that may come from three different suppliers at three different prices, with stock that moves several times a day. That's the mechanic we unpack here.

Cross-docking, classic stock, dropshipping: the right word for the right practice

Three models coexist, and they get mixed up constantly. The difference comes down to one thing: who holds the product, and where it passes through.

Classic stock stays the fastest for the customer, but it carries a cost people underestimate: according to the Institute for Supply Management, inventory carrying cost runs around 20 to 30% of inventory value per year. Tying up a 200,000-reference catalog at that rate makes no sense if half of it sells once a quarter.

Cross-docking keeps you in control of the shipment (you pack, label, inspect) while removing the holding cost. Dropshipping goes further: the supplier ships directly, and you never touch the product. Plenty of merchants use both, depending on the product and the supplier. The market reflects it: Grand View Research projects the global dropshipping market from 583.5 billion dollars in 2026 to 2,180.8 billion in 2033, a 20.7% compound annual growth rate.

Why more e-merchants sell without stocking

Because selling has shifted toward assortments nobody can physically warehouse. French e-commerce reached 196.4 billion euros in 2025 (+7% year over year, per the FEVAD), and much of it flows through marketplaces where catalog depth makes the sale. On Amazon, third-party sellers accounted for 62% of units sold in the fourth quarter of 2024 (Marketplace Pulse): that many merchants, most of them selling stock they don't hold.

The logic is easy to state, hard to run. If you can expose your suppliers' stock as if it were your own, your sellable catalog is no longer capped by the size of your warehouse. You move from an "own stock only" catalog to an "own stock + supplier stock" one. One of our clients, an auto body parts distributor, sells 200,000 to 500,000 references this way without stocking a single one: nearly all of its orders ship via cross-dock.

The hard part isn't the concept. It's everything between the customer's order and the parcel leaving: choosing the right supplier, creating the purchase order, receiving without errors, shipping fast. Done by hand, it collapses at the first volume spike. That's where the tooling earns its place.

How cross-docking works in e-commerce, step by step

E-commerce cross-docking is a chain of six links, and each one has to run without manual reconciliation to hold the volume:

1. Supplier stock synchronization. myFulfillment imports each supplier's stock and prices several times a day, via URL, FTP, or a dedicated crawler. It's the foundation: without fresh data, everything downstream is wrong.
2. Extended sellable stock. The stock pushed to your sales channels combines your own stock with your suppliers', carrying the right delivery message: "ships today" for what you hold, "ships in 72h" for what will transit.
3. Supplier selection. The same product (say a black bumper for a specific model) can come from several suppliers. On the order, the system ranks suppliers on stock, purchase price, and lead time, and proposes the primary supplier. The system proposes; the buyer keeps the final call.
4. Purchase order creation. Once the supplier is validated, the purchase order is generated automatically and sent by email, FTP file, or API call. Nobody watches a threshold by hand to place the order anymore.
5. Receiving. The product arrives. We treat this step below, because it's where most setups stall.
6. Transit and shipping. For a mono order, the received product goes straight to a packing station where the system matches it to the customer order. The operator packs, labels, and the parcel leaves.

The result at the auto parts distributor: 100% of orders processed via cross-dock, with no manual re-keying, and a transit time measured in minutes for mono orders where it used to take hours.

The trap nobody sees coming: receiving

Receiving is the link that breaks first, and almost nobody anticipates it. Here's why: your suppliers don't deliver one order at a time. They bundle several of your purchase orders into one box, one truck. With classic order-by-order receiving, the operator has to guess which product belongs to which PO. On a handful of orders, fine. On hundreds a day, it's the bottleneck.

The answer we built is called free receiving. The operator scans everything that arrives from the supplier without tying it to any purchase order. The system creates a bulk receiving record. Once scanning is done, an allocation matrix shows received products in columns and pending purchase orders in rows, and proposes an automatic assignment to the oldest orders. The buyer confirms or adjusts. That's it.

This detail changes the nature of the work. You go from order-by-order receiving to bulk receiving where the buyer only validates the matrix, and you remove the manual reconciliation that wasted the most time. It's exactly the kind of mechanic you can't guess until you've watched a bulk supplier delivery block an entire warehouse on a Monday morning.

Mono or multi: why one order ships in minutes and the other waits

Not all orders are equal, and treating them the same is a mistake. myFulfillment classifies each purchase order into two families, based on whether a single product is enough to fulfill the customer order:

• Mono: one product, from one supplier, fulfills the customer order. On receipt, the product goes to packing and ships. Ultra-short path.
• Multi: the customer order is waiting on several items. The received product is set aside, and the order is only released once everything has arrived.

Without that split, a complete order ready to leave would sit behind another still waiting on an item. With it, simple orders, often the majority, ship within minutes, and only genuinely incomplete orders wait.

Cross-docking or dropship: where one ends and the other begins

The line is clean once you have the right picture. In cross-dock, the product passes through you: you receive it, pack it, ship it. In dropship, the supplier ships directly to the end customer, and you never see it.

Dropship needs one more piece: a way for the supplier to confirm the shipment and send back the tracking number. That's the job of the supplier portal. The supplier receives the purchase order, which carries the end customer's address and a packing slip in your brand so the parcel arrives under your identity. It confirms the shipment, enters the tracking, and the information flows back automatically to the marketplace. The merchant no longer has to step in.

The two models don't exclude each other. On our side, roughly 50% of myFulfillment clients run cross-dock and 10 to 15% run dropship, often the same ones, deciding product by product, supplier by supplier.

What we see at Boostmyshop

These figures come from our own deployments, not from a market average. They say something simple: cross-dock isn't an edge case reserved for marketplace pure players. It has become a common operating pattern, and its tipping point is neither the concept nor the sale. It's automating receiving and purchasing.

Cross-docking: who it's for, and when to avoid it

Cross-docking is for you if you sell a catalog wider than what you can stock, if your suppliers are reliable on stock data, and if you accept showing a delivery time of around 72h on part of the assortment. It's the model for very large-catalog distributors: spare parts, electronic components, industrial MRO, plumbing and HVAC, any marketplace seller looking to expand without expanding the warehouse.

It's a worse fit when your promise rests on same-day shipping end to end, when your suppliers can't expose reliable stock, or when your margins can't absorb buying on order. In those cases, own stock on your best movers (where reliable replenishment forecasting makes the difference), topped up with cross-dock on the long tail, is often the right compromise.

A last word on the numbers you read everywhere. Stockouts are estimated to cost retailers more than 1,200 billion dollars a year (IHL Group estimate). That's an order of magnitude of the problem, not a promise: no tool recovers "half of those sales." Cross-dock acts on one precise cause: the stock you can't afford to carry.

In short

Cross-docking lets you sell without stocking by transiting the product from supplier to customer. The easy part is explaining it. The part that makes the difference is the mechanic: syncing supplier stock several times a day, selecting the right supplier, generating purchase orders on their own, and above all receiving in bulk without manual reconciliation. That's where the hours are won. Or where Monday mornings are lost.

To see how this chain is configured end to end, myFulfillment procurement covers supplier selection, purchase orders, and receiving, and our procurement masterclass shows it running live. Two concrete cases go further: multi-supplier cross-dock for an auto parts distributor with 500K references and supplier-side cross-dock and dropship automation.

Sources

• Maersk, Cross-docking (Logistics Explained), 2024.
• Institute for Supply Management, The Monthly Metric: Inventory Carrying Cost, 2022.
• Grand View Research, Dropshipping Market, 2026.
• Marketplace Pulse, Amazon third-party seller share, 2025.
• FEVAD, Bilan du e-commerce en France 2025, 2026.
• Eurostat, E-commerce statistics, 2024 data.
• IHL Group, Inventory Distortion (cited by Mirakl), 2025.
• Academic: Van Belle et al., Cross-docking: State of the art, Omega, 2012; Boysen et al., Cross dock scheduling, Omega, 2009.