What is EDI? A Working Guide for Apparel Brands

A wholesale operations manager at a $15M apparel brand opens her Monday inbox and finds a chargeback notice from Nordstrom: $4,200 for an ASN that arrived four hours after the carton hit the dock. The carton was on time. The 856 transaction was not. Two days later, a separate $7,800 deduction lands from Macy's because the SSCC label printed by the 3PL had the wrong check digit, and pallet receiving sent the load to a manual exception queue. Neither failure was visible in the order management system, because EDI was running in a parallel middleware tool that nobody on the operations team owned. This is what bad EDI looks like at apparel scale, and it shows up first as margin erosion long before it shows up as a system problem.

What is Electronic Data Interchange (EDI)?

Electronic Data Interchange (EDI) is a standardized way for trading partners to exchange structured business documents (purchase orders, order acknowledgments, advance shipping notices, invoices, remittance advice) directly between systems, without email, PDFs, or human re-keying. The documents follow a published standard (most commonly ANSI X12 in North America, EDIFACT in Europe) and travel over a secure transport such as AS2, SFTP, or a value-added network known as a VAN.

For an apparel brand, EDI is not a nice-to-have integration project. It is the operating contract between you and every wholesale or department-store retailer you sell to. Nordstrom, Macy's, Saks, Bloomingdale's, Belk, JCPenney, and Kohl's all require EDI for wholesale and dropship programs. Each of them publishes a routing guide that defines exactly which EDI transactions are mandatory, what carton-label formats are accepted, what timing windows apply, and what the chargeback schedule looks like when you miss any of it.

From building the Shopify connector and the EDI translation layer, the pattern I see again and again is that brands treat EDI as transport, when retailers treat it as a fulfillment workflow. That mismatch is where margin leaks.

Which EDI transactions actually matter for apparel?

Apparel brands deal with a small, predictable set of EDI transactions. Knowing what each one carries is the difference between debugging a real problem in 20 minutes and chasing a transport-level red herring for a day.

• EDI 850, Purchase Order. The retailer's order. Lines, quantities, ship windows, ship-to and bill-to, allowances, markdown terms, packing and labeling instructions.
• EDI 855, Purchase Order Acknowledgment. Your formal answer: accepted, accepted with changes, rejected. Most retailers require an 855 within 24 to 48 hours of receiving the 850.
• EDI 856, Advance Shipping Notice (ASN). The single most expensive transaction to get wrong. The 856 declares what is in the shipment, how it is packed, and what carton label (SSCC-18) corresponds to which contents, hours before the truck reaches the retailer's DC.
• EDI 810, Invoice. Your bill to the retailer. Must match the 856 exactly on units and pricing or it gets parked.
• EDI 940 / 945, Warehouse Shipping Order and Warehouse Shipping Advice. The handshake between an ERP and a 3PL or warehouse: 940 says "ship this," 945 confirms what actually went out.
• EDI 820, Payment Order / Remittance Advice. The retailer's record of what they paid you and, importantly, what they deducted and why. This is where chargebacks land.
• EDI 846, Inventory Inquiry / Advice. Used heavily in dropship to publish on-hand stock to the retailer's site so they only sell what you can actually ship.
• EDI 997, Functional Acknowledgment. A receipt for any inbound or outbound EDI document. If a retailer never returns a 997 for your 856, the ASN is functionally lost.

Every major North American apparel retailer runs ANSI X12 over AS2 or SFTP, sometimes through a VAN such as SPS, TrueCommerce, or OpenText. European retailers and most marketplaces in the EU and UK lean on EDIFACT. The transport matters less than people think. The transaction quality and timing are what produce or destroy margin.

Why do apparel-EDI failures cost so much money?

Retailer routing guides are dense, version-controlled documents that rewrite themselves every season. They specify timing windows in hours not days, carton-label formats down to the font size and position, ASN structure down to whether you nest packs inside cases inside pallets, and a chargeback schedule that compounds across categories. A single shipment can absorb multiple deductions at once.

The defensible ranges I see across apparel routing guides:

• Late ASN: $250 to $1,000 per occurrence. Some retailers also assess a percentage on top.
• ASN-physical mismatch (the 856 says one carton has a quantity, the carton actually has a different quantity): 1 to 3 percent of shipment value, sometimes higher on flagship programs.
• Wrong carton-label format (bad SSCC, missing GS1-128, label in the wrong corner of the carton): $100 to $500 per carton.
• Mis-pack (style or size in the wrong carton, prepack ratios off): 5 to 25 percent of shipment value.

For a $15M apparel brand running wholesale into three or four department stores, retailer chargebacks above 1 percent of wholesale revenue are common when EDI is bolted on. A clean program runs under 0.5 percent. The delta is real money, and it is structural, not seasonal.

From the integration incidents we have triaged this year, the most expensive ASN failures almost never come from the AS2 layer. They come from the warehouse cutting a shipment that does not match what the EDI tool already declared, because the warehouse is operating from one source of truth and the EDI tool is operating from another. That is a fulfillment workflow problem dressed up as a transport problem.

Where does EDI sit in the 6 Breakpoints?

EDI is a fingerprint workflow for two of the six places apparel operations break under growth, mapped in the 6 Breakpoints framework:

• BP4, order flow becomes harder to trust. The 850 and 855 live here. When wholesale POs land in a separate EDI tool from DTC and marketplace orders, the operations team loses a single view of demand. Allocation gets political. Customer service tells the retailer one ship date and the warehouse hits a different one.
• BP5, warehouse execution gets less predictable. The 856, the 940/945, and SSCC labels live here. Receiving, putaway, pick, pack, and ship all converge at the moment an ASN gets generated, and any drift between what the system thinks shipped and what actually shipped is the chargeback.

The framework's point is that these breakpoints reinforce each other. BP4 corrupts BP5, because if the order data is wrong, the ASN built on top of it is wrong. BP5 corrupts BP3 (inventory truth), because every mis-shipped carton creates a phantom inventory adjustment that nobody trusts. By the time it reaches BP6 reporting, finance is reconciling deductions instead of running the business.

Wholesale EDI vs dropship EDI: why the workflow is different

Brands often build their EDI program around traditional wholesale, then add dropship and discover that nothing transfers cleanly. The transactions look the same on paper. The operational meaning is not.

Wholesale EDI is bulk and predictable. The retailer issues an 850 for a season's buy, you acknowledge with an 855, ship to the retailer's DC across a defined window, fire one 856 per shipment with carton-level detail, invoice with one 810. Volume is concentrated. The chargeback exposure is per-shipment.

Dropship EDI is the opposite. The retailer publishes your products on their site. Each consumer order produces a small 850 (often one or two units), you acknowledge with an 855, ship the order directly to the consumer with a parcel label that matches the retailer's brand, fire an 856 against the parcel, invoice with an 810. Volume is highly fragmented. The chargeback exposure is per-order, and timing windows are tighter, often "ship within 24 hours of order" with the 856 due within an hour of label creation. You also need to keep an EDI 846 inventory feed live so the retailer's site only sells what you actually have.

A wholesale-built EDI tool can usually be coerced into doing dropship, but the operational implications are very different. Dropship lives or dies on real-time inventory accuracy, parcel-label automation, and a 24/7 exception queue. Treating it as a wholesale variant is how brands end up with $200,000 in dropship chargebacks against $1.4M in dropship revenue.

Native EDI vs bolt-on middleware: the real economics

The decision most growing apparel brands eventually face is whether EDI lives inside the ERP or as a separate vendor. Both can technically work. The economics diverge fast as the retailer count grows.

Bolt-on EDI middleware (SPS Commerce, TrueCommerce, DiCentral, and similar) gives you a managed translator. They map the retailer's spec to a canonical schema, you map the canonical schema to your ERP. They charge per trading partner, per transaction, per map, plus support. For one or two retailers it is the fastest way to be live. By retailer five or six, the cost stack reads like a second salary, and every retailer routing-guide change creates a coordination loop between the middleware vendor, your IT team, and your operations team.

Native EDI inside the ERP means the order arrives directly into the order record that the warehouse, allocation, and finance teams already use. The 855 fires automatically off the order's allocation result. The 856 is generated from the actual pick-pack-ship event, not from a separate declaration. Inventory commits update live so the 846 reflects reality. The cost stack is per-retailer onboarding and ongoing routing-guide maintenance, not per-transaction.

The defensible position: bolt-on EDI middleware is a tax you pay for every retailer you onboard. Native EDI scales; middleware does not. If your wholesale program is going to grow past three or four major retailers and you are adding dropship, the native model wins on cost and on chargeback rate. EDI built into an apparel ERP is the version of this that matches how apparel brands actually run.

How does EDI connect to the 3PL and warehouse?

Most apparel brands at $15M to $100M wholesale revenue do not run their own warehouse. They run with a fashion-specialist 3PL such as Bergen Logistics, which carries native retailer-EDI compliance for the major North American department stores out of the box, or partners that handle similar retailer compliance regionally. The handoff is the EDI 940 / 945 pair: the ERP sends a 940 telling the 3PL what to ship and to what retailer routing guide, and the 3PL returns a 945 with the actual carton manifest, which becomes the data source for the 856 ASN to the retailer.

The reason brands choose 3PLs with native retailer-EDI compliance is that the 3PL has already absorbed the carton-labeling rules, prepack logic, and SSCC schemes for Nordstrom, Macy's, Saks, Bloomingdale's, and the rest. The brand's job becomes coordination of POs, ASNs, and inventory truth, not running a routing-guide compliance team. When an EDI program looks chronically expensive, it is often because the brand is trying to do compliance work the 3PL is already doing, in parallel, in a separate tool.

For brands selling wholesale into Australia and New Zealand through partners like Lufema, the same pattern holds: the right operational layer combines a B2B portal for direct buyers with retailer-grade EDI for the bigger accounts, and the brand keeps one record of every order regardless of channel.

What this means for an apparel operations team

EDI is not a back-office IT topic. It is an operations topic that happens to need a translator. The teams that get retailer chargeback rates under 0.5 percent share four behaviors:

1. One record per order, regardless of source. An 850 from Nordstrom and an order from the brand's Shopify store both land in the same order management system. Allocation, inventory commits, and fulfillment events all update the same record.
2. The ASN is generated from physical pick-pack-ship data, not declared up front. The 856 reflects what actually went into the carton, not what the system hopes did. This single change eliminates most ASN-physical mismatch chargebacks.
3. Routing-guide changes are tracked and tested before they go live. When Macy's publishes a v6.2 update, somebody on the team owns reading it, mapping it, and running a controlled test ASN. Brands that miss this step get blindsided every season.
4. Chargebacks are reconciled weekly to the dollar. The 820 remittance advice gets parsed against expected payments, deductions get categorized by reason code, and the operational team sees the trend before the CFO does.

A clean apparel-EDI program does not just lower retailer chargebacks. It frees a wholesale operations manager from spending 6 to 9 hours a week reconciling inventory across Shopify, the 3PL, and wholesale, and it cuts the 2 to 3 percent oversell rate that most $5M to $50M apparel brands run during peak. The $15M brand does not need a bigger EDI tool. It needs the EDI to live inside the system that already owns the order.

What are the real challenges of running EDI?

Even with native EDI, the program needs operational care. The recurring issues:

• Routing-guide drift. Retailers update their guides multiple times a year. A label-position change of half an inch can trigger chargebacks across an entire season's shipments.
• Trading-partner onboarding. Each new retailer wants a test cycle: 850/855 round-trip, then a sample 856, then a controlled live shipment. Skipping the test cycle is how brands ship a season's first batch with broken SSCCs.
• Functional acknowledgment monitoring. If 997s stop coming back, EDI is failing silently. The right setup alerts the operations team when 997s do not return inside the expected window, not the next morning.
• Mapping ownership. Generic EDI vendors don't understand apparel. They treat ASN as a transport problem when it is a fulfillment workflow problem. Whoever owns the maps needs to understand prepacks, ratio packs, and SKU-vs-style hierarchies, not just X12 segments.

How should an apparel brand choose an EDI approach?

The criteria that hold up over time:

• Does the EDI tool share a record with order management, inventory, and the warehouse handoff? If not, you are buying a sync seam.
• Does the ASN get built from actual pick-pack-ship events? If the ASN is declared at the point the order is dropped to the 3PL, mismatch chargebacks are baked in.
• Is the 3PL already retailer-EDI compliant? Bergen Logistics and similar fashion-specialist 3PLs absorb most of the routing-guide labor. Generic 3PLs do not.
• How does the cost scale with retailer count? Per-trading-partner fees compound fast. Per-transaction fees compound faster on dropship.
• Who owns routing-guide updates? The answer should be a named person on the operations team, not a vendor support ticket.

If retailer chargebacks already exceed 1 percent of wholesale revenue, the EDI integration is the problem, not the warehouse. Fixing it usually means moving EDI inside the system that already owns the order, not adding another bolt-on tool on top.