Shared Fulfillment for Fashion: Replace Dedicated Capacity With a SKU-to-Promise Rule

Dedicated fulfillment capacity can feel like control. The building, labor plan, and operating routines belong to one brand. But when order volume is seasonal, channel mix changes quickly, and customers are spread across the country, that control can become a fixed-cost constraint.
For Love & Lemons recently made that trade-off explicit. The fashion brand moved U.S. fulfillment from a dedicated California operation to a multi-client facility in Terrell, Texas. The important lesson is not simply that shared warehouses are cheaper. It is that replacing dedicated capacity works only when every SKU is governed by a clear customer-promise rule.
The case for shared capacity is measurableβ
Supply Chain Dive reports that direct-to-consumer orders represent roughly 60% to 70% of For Love & Lemons' sales, with wholesale accounting for the balance. Its former operation tied cost and capacity to a California site supported entirely by the brand's own volume. During major sales, orders could remain backlogged for weeks.
The company began shipping from Cart.com's Texas facility in July 2026. A more central location improved access to East Coast customers, while shared labor and infrastructure allowed capacity to flex with demand. The first meaningful stress test came in October: 75% of a collaboration collection sold out within 48 hours, and the operation shipped 1,200 orders in 24 hours without adding staff.
Those figures illustrate the upside, but they do not prove that every brand should close a dedicated site. A move succeeds when it improves customer promises and unit economics without creating inventory contention or channel conflict.
Build the decision from four data setsβ
Before changing the fulfillment model, assemble at least 12 months of order-line history and test four dimensions.
First, profile orders: lines per order, units per line, basket value, return rate, gift or packaging requirements, and the share needing same-day release. A warehouse designed for simple single-line parcels may struggle with wholesale cartons or elaborate fashion presentation.
Second, profile SKUs. Record dimensions, velocity, seasonality, size and color variants, replenishment lead time, and special handling. Fashion inventory is especially vulnerable to fragmented availability: having the dress but not the size the customer ordered is not useful stock.
Third, map demand by geography and promised service. Compare actual transit days from the current and proposed nodes by ZIP code, not by broad region. A central site may improve national averages while making a high-value West Coast segment slower.
Fourth, model peaks by hour and day. Monthly averages hide the labor problem. The relevant question is whether the shared site can absorb a launch-day wave while other clients are also peaking. For Love & Lemons' 1,200-order day without extra staffing is a useful proof point because it tests burst capacity, not just storage.
Replace space ownership with a SKU-to-promise ruleβ
In a dedicated building, teams often rely on informal priority decisions. Shared fulfillment requires explicit logic. For each order, the allocation engine should evaluate:
- whether every required SKU is available and released for that channel;
- the customer's promised delivery date and destination;
- pick, pack, and carrier cutoff times at the facility;
- the service and cost of eligible carrier options;
- wholesale routing-guide or marketplace compliance rules; and
- inventory needed for already committed orders and upcoming launches.
The rule should allocate stock only when the selected facility and carrier can still meet the promise. If the parcel cutoff has passed, showing βavailableβ inventory to a customer is misleading unless a later delivery date is presented. If a scarce SKU is reserved for a wholesale launch, the DTC channel should not consume it merely because that order arrived first.
This is where fewer operational handoffs can help. SupplyChainBrain notes that separate warehousing, transloading, drayage, linehaul, and fulfillment providers create additional chokepoints in information flow, scheduling, and accountability. A shared provider does not eliminate risk, but a single inventory-to-shipment workflow makes ownership easier to define.
Score the operation beyond average ship timeβ
Averages can conceal the orders that damage customer trust. Review performance by SKU class, sales channel, customer promise, and demand wave using four groups of measures.
Promise accuracy should include orders shipped by cutoff, delivered on time, and promises changed after checkout. Measure the original commitment, not a date quietly revised after the order enters the warehouse.
Inventory contention should track allocation failures, split shipments, substitutions, and stock reserved by one channel while another loses a sale. Compare system inventory with pickable inventory; damaged, held, or misplaced units cannot fulfill a promise.
Pick priority should reveal queue age, launch-order backlog, and whether wholesale compliance work displaces DTC orders. Define escalation thresholds before a large release, including who can override normal sequencing.
Cost per shipped order should combine storage, pick and pack, packaging, transportation, accessorials, returns processing, and compliance penalties. Shared capacity may reduce fixed storage and labor exposure, but a cheaper warehouse is not cheaper if split shipments or premium carrier upgrades rise.
Connect fulfillment allocation to transportation executionβ
Warehouse and transportation decisions cannot operate on separate clocks. The warehouse may release an order at 4:55 p.m., but if the appropriate carrier's final trailer closed at 4:30, the customer has already lost a day.
A TMS should receive the order promise, ship-from node, carton estimate, and ready time before allocation is final. It can test eligible services against carrier cutoffs and transit commitments, return the feasible options, and then retain the selected plan. When inventory, labor, or carrier capacity changes, the system should flag affected promises and present recovery choices rather than waiting for a missed scan.
The same record should connect actual pick completion, manifesting, carrier acceptance, cost, and delivery. That makes it possible to distinguish a warehouse delay from a carrier miss and to improve the rule with real outcomes.
Shared fulfillment is not a surrender of control. Done properly, it replaces ownership of idle space with better control over decisions. The winning design assigns every SKU to a channel, facility, and carrier only when that combination can protect the customer promise at an acceptable delivered cost.
CXTMS connects inventory-ready events, carrier options, cutoffs, shipment costs, milestones, and customer commitments in one workflow. Request a CXTMS demo to see how your team can turn shared fulfillment capacity into reliable delivery promises.


