5 min read
In brief
- China direct has no minimum order and the lowest cash exposure; a local warehouse buys speed and lower refunds at the cost of capital tied up in stock.
- In our worked example, importing 300 units cuts landed unit cost from $6.80 to $5.47 — a $1.33 saving that takes roughly 45 sales a month just to cover storage.
- The decision is sequencing, not ideology: test demand with direct shipping, then move the products that prove themselves into local stock.
What the question is actually asking
The debate is usually framed as quality versus price. That is the least important part. The variables that decide whether a store survives are delivery time, cash cycle, and whether your published delivery promise is true.
Direct shipping from an overseas supplier costs you nothing until an order arrives. A local warehouse costs you money months before the first sale, and in exchange it removes days from the customer's wait. Those two facts explain almost every difference between the models.
The three models, honestly
| Model | Cash out before launch | Typical delivery to customer | Main risk |
|---|---|---|---|
| China direct (agent or supplier) | None — pay per order | Weeks, not days; varies by carrier and season | Slow delivery drives refunds, chargebacks and inaccurate delivery promises |
| Bulk import into a local 3PL | Stock, freight, duties and storage, paid upfront | Days inside your own market | Dead stock if the product does not sell; cash locked in inventory |
| Domestic dropship (local supplier) | None — pay per order, higher unit cost | Days | Thin margin and a supplier who can sell to your competitor tomorrow |
The arithmetic of a bulk buy
Take a product that costs $4.20 per unit from the supplier. Shipped per order through an agent at $2.60, it lands at $6.80 with no minimum order.
Buy 300 units instead and pay $380 in freight: 300 × $4.20 = $1,260, plus $380, is $1,640 — $5.47 per unit landed. The saving is $1.33 per unit, and it costs $1,640 of cash plus a lead time before you can sell anything.
| Measure | China direct | Bulk + local 3PL |
|---|---|---|
| Landed product cost per unit | $6.80 (incl. agent shipping) | $5.47 (incl. freight, 300 units) |
| Cash committed before first sale | $0 | $1,640 |
| Per-order fulfilment cost | Included above | $6.50 (pick, pack and domestic shipping) |
| Saving on goods | — | $1.33 per unit |
| Units needed per month to cover $60 storage | — | ≈45 |
At 300 units sold the bulk route saves $400. At 1,000 units it saves about $1,333. Both are real money, and both are smaller than the cash that was tied up to get them — which is why the decision depends on how fast the product sells, not on the unit price alone.
Note the last row of that table: the local 3PL charges $2.40 to pick and pack plus $4.10 to ship domestically, $6.50 per order, against $2.60 of agent shipping. Faster delivery has its own price before you count storage.
When local actually wins
- The product sells predictably — enough orders per month that the saving on goods exceeds storage and the cash is not needed elsewhere.
- The customer segment is impatient or repeat-buying: subscription, replacement parts, anything where a second purchase depends on the first arriving quickly.
- Refunds and chargebacks are already a material cost line. Slow delivery is a direct cause of both.
- You are advertising in a market where the published delivery window is part of the competitive offer, not a footnote.
And it loses when the product is unproven. Buying 300 units of something no one has bought yet converts a marketing experiment into an inventory problem.
Delivery time is a policy variable, not a comfort feature
Two systems read your delivery promise as a truth claim. Payment processors treat a high chargeback rate as a signal that the customer did not receive what was described. Merchant Center and the ad platforms treat shipping and returns accuracy as a trust signal, and a mismatch between what you promise and what the checkout or the carrier says is the kind of thing that starts an account review.
An honest long window converts worse than a dishonest short one and keeps the account alive. That trade is not close, and it is the single most common way a fulfilment shortcut becomes an existential problem.
Duties, tax and the part that changes
Importing changes your obligations in ways per-order shipping does not. Duty thresholds, VAT and import regimes differ by destination and are revised with some frequency — including the de minimis arrangements many dropshippers relied on.
We are not customs advisers and will not publish a rate table that could be stale within a quarter. Confirm the current position with a broker for the specific product and destination before you commit capital, and price the duty into the landed cost rather than discovering it on the invoice.
The question that decides it
Not “which is cheaper per unit”, but “how many units will I sell before this stock stops being an asset”. If the answer is fewer than the minimum order, buy direct and pay the premium for optionality.
How we would sequence it
- 1
Prove demand with the lowest commitment
Ship direct per order while you learn whether the product converts and at what cost per purchase. Accept slower delivery and say so honestly on the site.
- 2
Watch the refund and repeat signals
If refunds are driven by delivery time, or if customers buy twice, the calculation tilts towards local stock before the unit price does.
- 3
Import only what you have evidence for
Size the first bulk order to weeks of proven sales, not to the supplier's best price break. The second order is where the margin lives.
- 4
Keep one direct-ship channel open
New products and long-tail variants stay direct until they earn their place in the warehouse.
If you are deciding between an import and staying direct, we can model it against your actual order volume and margin instead of a generic rule of thumb — and flag where the delivery promise is creating policy risk.