A real ASG warehouse sortation floor behind the article’s 3PL decision framework. By Janson Wang, Founder of ASG Dropshipping · October 1, 2026 · 20 min read
A SKU on your store has been selling steadily for months. Someone tells you it’s “ready for 3PL.” You ask at what number, and you get four different answers from four different people, none of whom agree with the other three.
That’s not because any of them is wrong.
It’s because no single number applies to every SKU, not as a platform rule, and not as one consistent thing the platforms themselves will tell you, their own pages give situational reference points, not a universal threshold.
This article doesn’t give you a sixth guess.
It gives you the five things worth checking on a SKU-by-SKU basis, and a way to test the move on one SKU before you bet the rest of the catalog on it.
Quick answer: what order volume means a mature SKU is ready to move from China-direct to a 3PL, FBA or WFS?
There isn’t one number that applies to every SKU, and the platforms’ own pages say so.
Walmart’s own help page states plainly that “there are no minimum requirements” for Walmart Fulfillment Services, then recommends at least 50 items with continual replenishment “to see the full value and GMV growth we hope to build with you.” Amazon’s Multi-Channel Fulfillment page gives no number at all, only the instruction to keep enough inventory on hand to cover demand across connected channels.
Amazon’s FBA page treats 40 items a month as a cost-effectiveness reference for the Professional plan, and separately floats “more than 10 orders a day” as an example point to consider a 3PL, neither one an enrollment rule.
A SKU is ready to leave China-direct fulfillment when it clears two gates, route eligibility and a known rollback path, and the remaining checks, demand stability, replenishment uncertainty, full landed cost and service requirement, support at least a limited pilot rather than ruling one out, tested on one SKU with a plan to walk it back.
Key takeaways
Amazon and Walmart’s own help pages don’t set a required minimum order volume for FBA, MCF or WFS. The numbers on those pages are cost-effectiveness references, setup recommendations and example growth paths, not enrollment rules.
Amazon’s 181-day aged-inventory surcharge is a storage-duration threshold, not a volume threshold. Confusing the two will make you read FBA’s fee page for a rule that isn’t on it.
China-direct, an independent 3PL (including ASG’s own overseas-warehouse option), a platform warehouse such as FBA or WFS, and a hybrid split of the two change who holds inventory, who makes the delivery promise, and how much control stays with you. None of the four is a universal winner.
Clearing the two hard gates makes a SKU eligible to consider, not automatically ready; the remaining variables can still say don’t pilot, gather more evidence, or limited pilot.
Move one SKU that clears both gates and scores well enough to pilot, keep China-direct live as the fallback, and judge the pilot by what you can count: oversell events, real cost against your estimate, inventory age, and whether delivery actually matched the promise, over one full replenishment cycle, not by how fast the batch sold out.
If your store sells on Amazon, Walmart or Shopify and a SKU has been moving at a steady pace on China-direct fulfillment, you’ve probably had this conversation.
Someone, a consultant, a forum post, a warehouse sales rep, tells you there’s a volume where it “makes sense” to stop shipping each order from China and start pre-stocking that SKU closer to the customer.
Then you ask a second person, and the number changes.
That’s not an accident of who you asked. Amazon’s and Walmart’s own fulfillment pages don’t publish a single order-volume minimum for entry.
What they publish instead are situational reference points that don’t agree with each other because they’re not answering the same question: Amazon’s FBA page frames 40 items a month as a cost-effectiveness reference for one specific plan, then separately suggests a seller might “expand with 3PL when you start receiving more than 10 orders a day”; Walmart’s WFS page states there are no minimum requirements at all, then recommends 50 items as a setup target for seeing the program’s full value.
A real reply on Shopify’s own community forum shows the same range inside a single answer: the same person put self-fulfillment at fewer than 10 to 20 orders a day and outsourcing at “dozens or hundreds” a day, two different situations being described, not two people disagreeing on one threshold.
So this article doesn’t add a fifth guess.
It walks through what “ready” actually has to mean for one SKU, and gives you a worksheet to score your SKU against five things, some checkable from your own order history, others only from a quote you’ll need to request or from what the pilot itself tells you, not from a number someone else picked.
Here’s the boundary, up front, so you know what this does and doesn’t cover.
The Question This Article Answers, and the Order-Count It Refuses to Give
Topic
Covered here
Not covered here
Decision in question
Whether and when to pre-stock a single, already-selling SKU in a destination-country warehouse
Whether to move your full catalog, or how to vet a China-based 3PL once you’ve already decided to stock there (ASG’s separate China 3PL warehouse checklist covers that step)
Returns and reverse logistics
Not covered in operational detail or costed per SKU. Amazon’s FBA page lists a “Returns processing” fee and a separate removal/disposal/liquidation fee; Walmart’s WFS page describes in-store and mail returns beyond its “seamless returns” meta-description line. None of the pages this article checked gives a full, itemized returns bill for a specific SKU. This article gives no cost comparison for returns
Confirm the actual return process and its cost directly with the warehouse or platform you’re evaluating before you commit a SKU
Where the numbers come from
Amazon’s and Walmart’s own current help pages, read on October 1, 2026, plus ASG’s internal logistics documentation for the China-direct and overseas-warehouse legs
Duty and tax outcomes for any specific country, or a guaranteed delivery time for any SKU
One more limit before the rest of this. When this article says what a platform’s page “states” or “recommends,” it means the words on that page as read on the date above.
It isn’t a claim about how Amazon or Walmart runs their warehouses internally, and prices, thresholds and eligibility rules on any platform’s page can change after this was written.
A real ASG sortation workstation where routing decisions become physical parcel movements. There are four shapes this can take for one SKU, and none is a universal winner.
Each moves a different piece: who holds inventory, who promises delivery time, and how much control stays with you.
China-direct is what you’re likely running today. The unit sits in a China-based warehouse until an order comes in, then ships per order.
ASG’s own logistics documentation describes a 5-to-12-day reference range for its core-market line-haul service, framed as a route-based range, not a fixed promise for every country, product or season.
You keep full control over packaging and SKU decisions and don’t tie up cash pre-stocking a batch abroad. What you give up is the ability to promise next-day or two-day delivery.
For the store-side setup behind that route, ASG’s guide to Shopify shipping zones and rates for China fulfillment explains how the checkout promise and the physical route fit together.
An independent 3PL in the destination country means you ship a batch ahead of demand into a third-party warehouse near your customers.
ASG operates one version itself: an overseas-warehouse option built for high-repurchase or high-volume SKUs, targeting a 1-to-3-day local-delivery reference figure once stock is in place, not a fixed promise, at the cost of needing that stock pre-positioned.
That’s one instance of this category, not the category itself; the same worksheet applies to ASG’s overseas warehouse and to any other third-party warehouse, without assuming either is a substitute for the other.
A platform warehouse, FBA or WFS, means the platform holds the stock and makes the delivery promise.
FBA-eligible products carry Amazon’s free two-day Prime shipping; WFS advertises delivery “in two days or less,” though neither promise is broken down by product weight or ship-to address on the pages this article checked, so confirm it for your specific SKU rather than assuming it applies uniformly (see the worked WFS example below).
Both have published eligibility rules (below), and both charge for the space your inventory occupies.
A hybrid split keeps some of a SKU’s inventory on China-direct and moves the rest to a 3PL or platform warehouse.
Shopify’s own help documentation confirms merchants can “fulfill orders yourself, use a fulfillment service, or combine both methods,” and that a single order can use “different fulfillment methods for different items in the same order.” Amazon’s Multi-Channel Fulfillment program is the platform-side version of the same idea: it lets FBA inventory fulfil orders placed on other sales channels, including your own storefront.
Four Ways to Fulfil a Mature SKU, and What Each One Actually Changes
Route
Who holds inventory
Who promises delivery time
What you keep control of
What this article hasn’t checked
China-direct
You, in China, until the order ships
You, per order, within a reference range
Packaging, SKU decisions, no pre-stocking commitment
—
Independent 3PL abroad (including ASG’s overseas-warehouse option)
The 3PL, once you’ve pre-stocked it
The 3PL, per its own published or agreed terms
Less than China-direct; you’re committed to a batch
Specific SLA terms vary by 3PL and weren’t tested here
Platform warehouse (FBA / WFS)
The platform
The platform, per its published delivery promise
Least; the platform sets eligibility and storage rules
Fee tables beyond what’s on the pages cited
Hybrid (split by SKU or by channel)
Split between you and a partner
Split, depending on which leg fulfils the order
A deliberate middle ground you design
Whether one order can be split across two fulfillers (see later in this article)
No route here outranks another. Platform warehouses buy speed at the cost of control and a storage bill; China-direct keeps control at the cost of speed.
The rest of this article helps you figure out which trade you want, for one SKU.
A product being prepared at an ASG China-based warehouse workbench. If a single order-count can’t tell you a SKU is ready, what can?
Five things, though not all from the same kind of data: some from your own order history, some from a quote you’ll need to request, and at least one you may only be able to test in the pilot itself.
The five readiness variables, one line each
Demand stability: whether this SKU’s sales over a full replenishment cycle fall in a range narrow enough that you’re willing to pre-stock a batch against it.Replenishment uncertainty: how long it takes, and how much that time varies, from placing a reorder to having stock received and available in the destination warehouse, customs included.Full landed and warehouse cost: the true per-unit cost once inbound freight, storage, pick-and-pack, and any surcharges are added, compared against what the same unit costs you on China-direct.Service requirement: whether this specific listing’s conversion actually depends on a fast delivery promise, or whether buyers are already accepting your current transit time.Rollback capacity: whether the SKU’s orders can switch back to China-direct inside the same store if the pilot doesn’t work, and whether the cost of any stranded inventory is known in advance.
Rollback capacity appears twice here: once as one of the five variables, and again below as a hard gate a SKU must clear before any pilot starts. That’s not a contradiction.
The other four variables answer an expected-value question, is the move worth it, is the timing right. Rollback asks something different: if you’re wrong, can you undo it, and at what cost.
A bad expected-value call costs some margin for a while.
An unrecoverable one can cost you the SKU, the channel relationship, or stranded inventory with nowhere to go, whether you pulled back or a platform changed your listing out from under you.
That gap in downside is why rollback gets its own pass-or-fail check instead of staying folded into a score.
Before any of the five variables matters, one thing has to be true first: the SKU has to be physically and legally eligible for the route you’re considering. Call this Gate 0 .
Walmart’s WFS eligibility rules, for instance, cap products at 500 lbs and 120-by-105-by-93 inches including packaging, require the product to be non-perishable and not need temperature control, and require inventory that’s already shipped from the U.S. or cleared through customs.
That’s a platform rule.
It’s a different thing from a real seller’s note on Shopify’s community forum that handmade, kitted or customized products often fit better kept in-house, a business-fit judgment about who should handle the product, not a warehouse eligibility rule, and it shouldn’t be scored as one.
Gate 0 isn’t a sixth variable to score, and passing it isn’t the same as being ready to pilot.
It’s a yes-or-no check that comes before scoring starts: a SKU that fails it can’t be helped by a good answer on any of the five, and a SKU that passes it still has to clear the rollback gate and hold up on the remaining variables before a pilot makes sense.
These five checks aren’t meant to be multiplied into one composite score.
A SKU can fail on one variable and still clear a pilot on a smaller scale, or clear four variables and fail the fifth in a way that rules the whole move out for now.
The worksheet below keeps them as five separate columns for exactly that reason.
The article’s decision model: two gates, five variables and three possible outcomes.
This is the worksheet itself. Copy it into your own spreadsheet, one row per SKU you’re considering.
Where you don’t have real data yet, an actual quote, a tested lead time, write “unknown” rather than filling in a guess; an honest unknown is more useful than a false “ready.”
The Readiness Worksheet: Scoring Each Variable From Your Own Order Data
Variable
Where to look in your own back office
What to write down
Looks like “not yet”
Looks like “ready to pilot”
Demand stability
Sales by week or by replenishment cycle, for this SKU alone
The range your sales have fallen in over the last full cycle or two
Sales swing widely with promotions or season, with no repeatable floor
Sales sit in a range narrow enough that you’d commit a batch against it
Replenishment uncertainty
Your supplier or agent’s quoted and actual lead times, including any customs step, through the stock being received and confirmed available for fulfillment, not just arrived
Demand during that full cycle, plus the longest and shortest lead times you’ve actually observed
Available stock and confirmed in-transit replenishment don’t cover demand through the next delivery, or your buffer can’t absorb the variation you’ve seen
Available stock and confirmed in-transit replenishment cover the cycle, and your buffer comfortably absorbs the longest lead time you’ve observed
Full landed cost
Your product cost, plus quotes for inbound freight, storage, pick-pack and any surcharges at the warehouse you’re evaluating
Your own per-unit estimate, built from the cost formula in the next section, with every line stated in the same unit
The destination-warehouse estimate comes out higher than China-direct once every line is added, at the same per-unit basis
The destination-warehouse estimate holds up against China-direct at the same per-unit basis, or the gap is one you’ve decided to pay for speed
Service requirement
Conversion rate or buyer messages tied to delivery speed on this specific listing, or an actual channel or customer commitment you already have to meet
Whether buyers are asking for faster delivery, abandoning at checkout over shipping time, or neither, and whether any channel or agreement already commits you to a delivery window
Buyers are accepting your current transit time without friction, and nothing commits you to a faster one
You have a real, comparable signal, a documented commitment, a pattern of checkout abandonment tied to shipping time, not just a hunch that faster delivery would change the outcome
Rollback capacity
Your store’s fulfillment settings, and what you know about the destination warehouse’s exit terms
Whether future orders can switch back to China-direct inside the same store, what happens to orders the warehouse has already accepted, and what happens to stock left there
You can’t say what happens to orders already accepted or to unsold stock if you stop, or whether the store can switch fulfillment method without relisting
You know all three answers and are comfortable with the worst case
Two hard gates sit above this table, checked before the four remaining columns matter. Gate 0, route eligibility, comes first: weight, dimensions, product category and customs status. Rollback capacity is the second gate, checked here at its minimum viable version: a SKU that can’t answer “how do we walk this back” doesn’t enter a pilot, no matter how it scores elsewhere.
Clearing both gates makes a SKU eligible to consider, not automatically ready.
The remaining variables, demand stability, replenishment uncertainty, full landed cost and service requirement, then point to one of three outcomes: don’t pilot yet, because the demand or the money doesn’t hold up; gather more evidence first; or run a limited pilot on this SKU.
A SKU that’s exitable but can’t be funded, or whose demand swings too widely to commit a batch against, doesn’t get waved through just because it cleared the gates.
When you total up a pilot’s results later, keep the denominators tied to one SKU and one batch: an oversell count is events divided by that SKU’s orders in the pilot window; a per-unit cost is the batch’s total cost divided by its unit count.
Don’t blend batches from different routes or seasons into one average, it hides the variation this worksheet is trying to surface.
Get ASG’s current reference numbers for your supply leg.
Filling in the “full landed cost” row means having real numbers for consolidated inbound freight, customs and lead time from China to your destination warehouse.
Send ASG the SKU, destination and typical order quantity, and ask for the current reference range for that leg, not a quote for the whole move.
Ask ASG for Your Numbers
A real ASG order workstation used to review order data and prepare outbound work. A worksheet score is only as good as the cost line behind it:
China-direct cost per unit = (product cost + processing and packaging + per-order line-haul shipping, route-based and typically a multi-day reference range, not a fixed promise + any per-shipment duties or taxes for your destination) ÷ the number of units in that order, plus returns handling, only if it applies to this unit (no source checked here gives a dollar figure; confirm with your own returns process)
Destination-warehouse cost per unit = product cost + consolidated inbound freight per unit + customs and duties per unit + receiving or prep fee + monthly storage (space held × months held, at your negotiated or quoted rate) + fulfillment fee per unit, confirmed to include outbound delivery to the customer or with that delivery added separately if it doesn’t + surcharges that actually apply (aged inventory, peak season) + rollback or disposal cost, only if it applies + returns handling (no source checked here gives a dollar figure; confirm directly with the warehouse or platform you’re evaluating)
Both formulas are now per unit, on purpose: the original per-order version for China-direct couldn’t be compared against a per-unit destination-warehouse figure once an order holds more than one unit.
Both also carry a returns line without a dollar figure attached, also on purpose, but that’s not the same as “unsourced” in the sense of “not addressed.” Amazon’s FBA page lists a “Returns processing” fee and a separate removal/disposal/liquidation fee; Walmart’s WFS page describes in-store and mail returns beyond its “seamless returns” meta-description line.
These platforms do address returns, just not with a per-SKU number you can copy into a formula.
Leaving the line out entirely would quietly suggest returns are free once a SKU moves off China-direct, and nothing here supports that.
Confirm the real number, and how it’s billed, before you commit, not after.
The fulfillment-fee line needs the same care: FBA’s and MCF’s published fulfillment fees already include picking, packing and shipping, so don’t add a separate delivery charge on top of them; an independent 3PL’s quote may or may not include that same end-mile delivery, so confirm which before you add it in.
The distinction that does the most damage to “flat rate” thinking: what’s a published platform fact, and what’s an input only you can supply.
The Full-Cost Comparison: The Formula, and Which Inputs Are Yours Versus the Platform’s
Cost line
Official fact from the platform’s page
Your own input
Storage
FBA charges monthly, based on the cubic feet your inventory occupies, with an aged-inventory surcharge starting past 181 days in storage
How many months you expect to hold this SKU’s batch
Per-unit fulfillment fee
FBA’s page doesn’t publish this rate here; it points sellers to Amazon’s Revenue Calculator instead
Your actual per-unit fee, pulled from that calculator for this SKU
WFS fees
Walmart states zero monthly or setup fees for WFS
Your inbound freight, customs and any per-unit fulfillment cost, none of which are zero
Product and freight
Not a platform fact; this is entirely your number
Product cost, consolidated inbound freight rate, customs and duties for your specific lane
Returns
Addressed qualitatively by the sources checked here, FBA’s “Returns processing” and removal/disposal fees, WFS’s in-store/mail returns, but no dollar figure for a specific SKU
Confirm the actual fee and process with the specific warehouse or platform before you rely on a number
A real seller on Shopify’s community forum put the general shape of this plainly: “the cheapest quote almost never ends up being the cheapest option,” once receiving fees, storage, packaging and peak-season surcharges are added to a quote that looked clean on its own.
That’s not a claim about any specific warehouse; it’s a reminder to price every line in both formulas above before comparing totals, not just the headline rate.
One number deserves a closer look because it’s easy to misread as a per-SKU discount.
Walmart states WFS costs “15% less than other marketplace fulfillment providers,” but the figure carries its own footnote: Walmart’s first-party data, orders fulfilled July 1, 2024 to June 30, 2025, measured as fulfillment cost per item, across Walmart’s own comparison sample, not a quote for your SKU.
Keep it as attributed background only, not a discount factor to plug into the cost-per-unit formula above; build that formula from your own inbound freight, customs and storage numbers instead.
If your storage months double, the storage line in the destination-warehouse formula grows while the China-direct line doesn’t move, the kind of shift a flat percentage can hide.
The two article formulas aligned on the same per-unit basis; no dollar figures are invented. If you’d rather see how a China-based fulfillment quote breaks down into its component fees before you compare it against a destination-warehouse quote, ASG has a separate breakdown of the hidden costs in a dropshipping agent’s quote .
Take one case: a SKU selling steadily on Walmart Marketplace through China-direct fulfillment.
Walmart’s WFS help page gives the clearest official facts of any platform covered here, so it’s worth walking through on its own.
Eligibility first.
WFS accepts products “up to 500 lbs” with maximum dimensions of “120″ x 105″ x 93″, including packaging,” non-perishable, no temperature control, and “shipped from the U.S. or cleared through customs” before arrival.
That last condition matters for a China-direct seller: your replenishment-uncertainty number has to include customs clearance time, not just transit time to the port.
On volume, Walmart’s own FAQ answers the question directly: asked whether there are inventory requirements or SKU minimums to ship with WFS, the page states “at this time, there are no minimum requirements,” then recommends “at least 50 items with continual inventory replenishment to see the full value and GMV growth we hope to build with you.” That’s a recommendation about getting value, not a gate you have to clear.
Fees are “zero monthly or set up fees,” and the publicly advertised delivery promise is “two days or less,” though WFS’s own page doesn’t break that promise down by product weight or ship-to address, so confirm it actually applies to a SKU at the top of that 500 lb range before you rely on it.
Run this SKU through the five variables.
On service requirement , the real question is whether Walmart buyers are actually asking for two-day delivery, or converting fine on your current timeline; Walmart’s own data ties the “Fulfilled by Walmart” tag to an average 50% GMV growth, its own aggregate figure, not a promise for your SKU.
On replenishment uncertainty , the “cleared through customs” eligibility rule means your lead-time number has to fold in clearance time before stock is available to fulfill from the WFS warehouse; for China-direct, clearance happens during each order’s own cross-border transit instead, it isn’t absent, it’s positioned differently.
On full cost , zero monthly fees doesn’t mean zero cost: storage, pick-pack and surcharges still apply, so plug WFS’s own numbers into the per-unit formula above rather than stopping at “no setup fee.” On rollback , no page here describes what happens to orders WFS has already accepted, or to stock left in a WFS warehouse if you stop sending it, so both of those boxes should read “unknown, ask Walmart.”
None of this settles “move it” or “don’t.” It tells you which of the five boxes this SKU fills in, and which ones still need an answer before you decide.
A real ASG consolidation area; it is not presented as a Walmart-operated facility. If you sell the same SKU across Amazon, Walmart and your own Shopify store, the question isn’t just “should this SKU move,” it’s “which order gets fulfilled by which route.”
Amazon’s Multi-Channel Fulfillment program is built for exactly this, and it’s specifically the off-Amazon path: orders placed on Amazon.com itself still fulfil through regular FBA, not through MCF.
Its own page describes using FBA inventory “to fulfill customer orders from other sales channels, including your own website,” backed by “100+ prebuilt and developer-friendly integrations,” naming Shopify and WooCommerce specifically, or a custom integration.
That connection has to be set up first, through one of those integrations, a custom build, or manually creating MCF orders in Seller Central; holding FBA inventory doesn’t connect anything automatically.
Delivery runs on two tiers: Expedited, “two business days,” and Standard, “three business days,” click to delivery.
New sellers get 25% off fulfillment costs on the first 100 units shipped through MCF, a limited, first-100-units incentive, not a standing rate.
The only volume guidance on MCF’s FAQ is a recommendation, not a number: “make sure you keep enough inventory on hand to cover demand across all your connected channels.”
On the Shopify side, the structural piece that makes routing possible is already built in, and it’s more specific than a general “combine methods” statement: Shopify’s help documentation confirms merchants can fulfill orders themselves, use a fulfillment service, or combine both, and that a single order can use different fulfillment methods for different line items inside it, meaning an order with two SKUs really can route one item through China-direct and the other through a 3PL.
What the documentation doesn’t spell out, on the pages this article checked, is how a fulfillment request behaves once it’s already been sent, and especially once it’s been accepted: switching a line item to a different fulfiller at that point may require cancelling the original request and waiting for that provider’s confirmation, not just changing a store setting.
Test that specific sequence on your own store before you rely on it for a live pilot.
Amazon MCF, WFS and China Routing for Omnichannel Sellers: Who Fulfils Which Order
Sales channel
Route
Who holds inventory
Who promises the delivery time
Who keeps inventory records in sync
What this article hasn’t confirmed
Amazon
FBA (direct)
Amazon
Amazon, per FBA’s own terms
Amazon’s own system
—
Walmart or Shopify orders, routed through your FBA inventory
MCF
Amazon
Amazon, on MCF’s Expedited/Standard tiers
Amazon’s MCF system, once you’ve connected that channel through an integration, a custom build, or manual order entry
Packaging and carrier conditions for MCF orders routed to another platform; this round didn’t reach Amazon’s full MCF FAQ to confirm them
Walmart.com orders
WFS
Walmart
Walmart, “two days or less”
Walmart’s own system
Handling of orders already accepted, and stranded inventory, if you stop
Your own site or other marketplaces, via Walmart’s Multichannel Solutions program
WFS Multichannel Solutions
Walmart
Walmart, per Multichannel Solutions terms; not confirmed identical to the Walmart.com promise above
Walmart’s own system
Eligibility, fees and service terms for this option specifically; this article only confirms it exists
Shopify (your own store)
China-direct
You
You, within a reference range
You
—
Shopify (your own store)
Independent 3PL or hybrid
Split, you and the 3PL
Split
You and the 3PL, by agreement
Whether an already-accepted order can be reassigned without the first fulfiller’s confirmation
Walmart’s WFS source material names this Multichannel Solutions option explicitly, fulfilling orders placed on “your website, Amazon, eBay, TikTok and more,” with unbranded packaging for those orders versus Walmart-branded packaging for Walmart.com orders.
That’s the platform confirming the option exists; it isn’t this article confirming the specific fees or SLA terms that would apply if you used it, those still need checking directly with Walmart.
The inventory-sync risk belongs here, not with volume.
Replies on Shopify’s own community forum warned that even a small inventory mismatch between systems can create overselling, a caution worth taking seriously once one SKU is tracked by more than one system, your store, plus a 3PL, plus MCF’s shared pool, plus WFS’s own system, depending on which routes you’re running.
That’s a reason to check your own sync setup before you rely on it, not a documented platform failure with a known cause; treat any specific sync behavior you haven’t tested yourself as unconfirmed.
An ASG scanning station where parcel handling and order records meet. Once a SKU clears Gate 0 and the rollback gate, the next question isn’t “should we migrate the catalog,” it’s “how do we test this on one SKU without betting the rest of it.”
Four rules for the pilot.
Move one SKU, or very few, that clears both gates and whose remaining variables point to “limited pilot,” not the SKU with the easiest numbers or the one that merely hasn’t failed anything yet.
Keep the China-direct leg live for that SKU while the pilot runs; Shopify’s own support for combining fulfillment methods inside one store makes this possible without relisting anything.
Measure the replenishment cycle itself, reorder to stock confirmed available at the destination warehouse, separately from how long that batch then takes to sell through; don’t collapse the two into one fixed day count picked in advance.
And only count things you can actually count.
Run a Reversible Pilot: What to Move First, What to Count, and the Signals That Stop or Expand It
Signal
Direction
What to count
Denominator
Unexplained oversell or stock mismatch
Stop
Oversell events during the pilot window
That SKU’s orders during the pilot window
Real per-unit cost versus your worksheet estimate
Stop if it exceeds your own margin buffer
Actual batch cost from the destination warehouse
That batch’s unit count
Inventory age approaching the surcharge line
Stop
Days in storage for the unsold portion of the batch
— (compare directly against the 181-day FBA aged-inventory threshold if that’s your route)
Replenishment worst case versus safety stock
Stop if it’s exceeded
Longest observed reorder-to-available time (received and confirmed fulfillable, not just arrived)
Your safety-stock coverage in days
Delivery, cost and records reconciling
Expand
Orders fulfilled within the promised window, batch cost within your estimate, and inventory records matching what’s on hand
That SKU’s order and batch count during the pilot
Not enough evidence yet, no stop signal, but delivery, cost or records still unconfirmed
Keep observing, don’t count as expand
—
—
A stop signal doesn’t mean the SKU failed forever.
Walk back future orders using the rollback path you confirmed before the pilot started, figure out which variable was wrong, and either fix the input or leave that SKU on China-direct.
If no stop signal has fired but you also can’t yet confirm delivery, cost or records, that’s not an expand signal either; keep observing rather than recording a pass.
An expand signal, delivery matching the promise, cost holding up, records reconciling, means the next move is adding one more SKU or opening the pipeline to a second channel through something like MCF.
Rollback has three separate parts, not one switch.
Future orders: switching a SKU’s routing back to China-direct happens inside your store’s own fulfillment settings, the mechanism Shopify’s documentation confirms supports combined and per-item fulfillment methods.
Orders the destination warehouse has already accepted: those may need to be cancelled, and that cancellation confirmed by the warehouse, before you ship the same units from China again; changing your store’s setting doesn’t by itself stop a shipment already in motion.
Stock left in the destination warehouse: this article can’t hand you a number for its disposition cost.
Confirm all three, not just the first, with the specific warehouse or platform before the pilot starts, not after.
A real ASG product-handling step that can remain active during a limited fulfillment pilot. A reversible pilot loop: incomplete evidence is not an expand signal. ASG sits in two places here, worth being precise about so neither gets overstated.
The first is the leg you’re probably already running: China-direct fulfillment.
ASG’s own logistics documentation describes this as a 5-to-12-day reference range for core markets, built for standard per-order dropshipping, not a fixed promise across every country, product and season.
Whatever you decide about moving a SKU abroad, this is the leg that stays live during the pilot as your fallback.
The second is the supply source behind any of the four routes, including the ones that don’t involve ASG’s own warehouses.
Whether a SKU ends up on China-direct, an independent 3PL, FBA, WFS or a hybrid split, the inventory going into that destination warehouse still has to be sourced, quality-checked and consolidated out of China.
That’s the part ASG operates directly: 4 warehouses across Shenzhen and Dongguan, 2,300+ verified factories, a catalog above 1.4M+ SKU, and 5,000+ sellers served, operating since 2019.
ASG’s standard published lead-time reference for this sourcing-and-ship flow is typically 1 to 3 days processing, 5 to 8 days delivery to the USA, UK and Europe, a reference sentence, not a guarantee, and it describes the same China-direct route as the 5-to-12-day core-market range above, not a separate leg.
The two figures come from different ASG documents and roughly line up, 6 to 11 days sits inside 5 to 12, but don’t substitute either one for the destination-warehouse replenishment time a 3PL, FBA or WFS pilot actually depends on; that number has to come from the warehouse you’re evaluating, not from either of these.
ASG also operates its own overseas-warehouse option, targeting a 1-to-3-day local-delivery reference figure, not a fixed promise, once a SKU is pre-stocked.
That’s one instance of the “independent 3PL” row above, evaluated against the same worksheet as any third-party warehouse, not a special case.
Moving a SKU to ASG’s overseas warehouse isn’t the same decision as moving it to FBA or WFS.
A few things ASG doesn’t do. It doesn’t answer for FBA’s or WFS’s inspection, performance or policy outcomes; that runs between you and the platform.
It doesn’t promise a fixed delivery date, only the reference range above.
And “helping prepare inventory to meet a destination warehouse’s inbound requirements” is a direction ASG can move in, not a documented service with a fixed scope; ask what’s included for your SKU and destination.
If you’ve already decided to move inventory to a China-based 3PL rather than the destination-country question covered here, ASG’s separate 15-question checklist for vetting a China 3PL warehouse picks up from there.
A real ASG quality-check bench for the China supply leg.
Keep your China-direct leg live while you pilot one SKU abroad.
Whichever of the four routes you’re testing, the supply side, sourcing, QC and consolidation out of China, still has to work while you run the pilot.
Tell ASG which SKU you’re testing and where it’s headed, and ask how the supply leg would stay steady during the test.
Talk to ASG About Your China-Direct Leg
Is there an official minimum order volume before FBA, MCF or WFS makes sense?
No single number that applies to every SKU.
Walmart’s own WFS FAQ states “there are no minimum requirements,” then recommends at least 50 items, its own wording, not a converted unit count, with continual replenishment to see the full value of the program.
Amazon’s MCF page gives no number at all, only an instruction to keep enough inventory on hand to cover demand across connected channels.
FBA’s “40 items or more per month” line is framed as a cost-effectiveness reference for the Professional plan, and FBA separately floats “more than 10 orders a day” as an example point to consider a 3PL; neither one is an enrollment rule, and neither is meant as a general-purpose threshold for other sellers’ SKUs.
What’s the difference between an order-volume threshold and an inventory-age threshold?
They’re triggered by different things. Amazon’s 181-day surcharge is triggered by how long a unit has sat in storage, not by how many units you’ve sold.
Reading it as a volume rule sends you looking for a number on FBA’s fee page that isn’t there.
Can one Shopify store keep some SKUs on China-direct and send others to a 3PL or FBA?
Yes: Shopify’s own help documentation confirms you can combine self-fulfillment with a fulfillment service, and that a single order can use different fulfillment methods for different line items, so an order with two SKUs really can route one through China-direct and the other through a 3PL.
What’s unconfirmed here is what happens once a fulfillment request for one of those items has already been sent and accepted, switching it to a different fulfiller at that point may need a cancellation confirmed by the original provider, not just a settings change.
Test that specific sequence on your own store before planning a live routing setup around it.
Should Walmart’s “15% cheaper” figure decide the move?
No.
It’s Walmart’s own first-party data, orders fulfilled July 1, 2024 to June 30, 2025, measured as fulfillment cost per item across Walmart’s own comparison sample, not a quote for your SKU and not a discount factor to apply to China-direct.
Keep it as attributed background, and build your own full-cost formula from your actual inbound freight, customs, storage and fulfillment-fee numbers instead.
How do I tell a failed pilot from one that’s just starting slowly, or from one that simply doesn’t have enough evidence yet?
Watch what’s actually countable.
An unexplained oversell or stock mismatch, a real per-unit cost past your own margin buffer, or unsold stock approaching an aged-inventory surcharge line are stop signals.
Delivery matching the promise, cost holding up and inventory records reconciling, checked against real orders, are expand signals.
If none of the stop signals have fired but you also don’t yet have delivery, cost and record data to check against, that’s not evidence of success, it’s not enough evidence yet; keep observing before you call it either way.
How this article was sourced. The platform pages above were read on October 1, 2026; figures and quotes are copied from those pages as read that date.
The Shopify Community thread is a single real conversation, quoted as written, not a statistical sample; the self-fulfillment-versus-3PL range it contains comes from one reply describing different situations, shown as evidence there’s no single agreed number, not as a threshold to adopt.
This article doesn’t test any third-party app, doesn’t cover returns or reverse logistics, and doesn’t independently verify Walmart’s or Amazon’s self-reported performance figures.
ASG data note. ASG figures in this article, including years of operation, warehouse count, factory count, catalog size, seller count and lead-time ranges for the China-direct and overseas-warehouse legs, come from ASG’s internal operating records and logistics documentation.
