The Real Cost of Buying Direct from China

A seller switches a $2.90 product to a factory quoting $2.35 and books the 19% saving into next year's plan. Twelve months later margin is flat. The order came with a 3,000-piece minimum against annual sales of 1,800, two rounds of samples that were charged for, an inspection he skipped because it cost more than it seemed worth, and roughly forty hours of his own evenings spent on messages across a seven-hour time difference.

Nothing in that story is unusual and nothing went badly wrong. The saving was real at the unit level and disappeared into lines that never made it onto the comparison. Working out whether direct buying pays is an arithmetic problem, and the arithmetic has nine terms rather than one.

Key Takeaways

• Unit price comparison structurally favours going direct, because the costs it excludes fall mostly on the direct route.

• The correct denominator is landed cost per saleable unit, which counts freight, duty, handling and units that arrive unsellable.

• Minimum order overhang is usually the single largest hidden cost, and it grows with the discount that tempted you.

• Inspection is cheap per unit at volume and uneconomic at small quantities, which is why small direct orders often go unchecked.

• Break-even depends on annual volume per product and on how many products you run, not on the size of the discount.

Why Unit Price Comparison Always Favours Direct

Compare two quotes and the factory number will nearly always be lower. That is not evidence of a better deal, because the comparison is measuring one route with its costs removed.

What the Quoted Price Excludes

A factory quote covers goods at a defined point in the journey and nothing else. Supplier vetting, sampling rounds, quality control, export coordination, freight booking and problem-solving all sit outside it. Buying through an intermediary bundles those into a price that therefore looks higher while covering more.

The comparison only becomes meaningful when both routes are priced to the same endpoint. That means quoting everything on the same Incoterm, then adding every downstream cost to whichever route does not already include it.

The Costs That Appear on Only One Route

Nine lines make up a complete comparison: unit price, freight and destination charges, duty and clearance, inspection, sampling and setup, minimum order overhang, cash tied up, defect and rework losses, and your own time. An intermediary price typically absorbs several of these. A factory price absorbs one.

Write them as a list before you request any quotes and fill each cell for both routes. The exercise takes an afternoon and it changes decisions, because the lines that decide the answer are rarely the ones buyers argue about.

Scope creep runs the other way too, and it is worth naming. Some intermediary quotes exclude inspection, or include only a carton count rather than a product check. Ask both routes exactly which of the nine lines their number covers, in writing, and the comparison stops being a guess about what each side assumed.

Expert Tip:Price your own hours before you start, and use a number you would actually accept. I use what an hour of my time earns in the part of the business that grows revenue, not a notional wage. Forty hours of supplier management at that rate is a real cost line and it belongs in the model. Buyers who leave their own time at zero always conclude that direct is cheaper, because they have removed the largest cost the direct route carries.

Start With Landed Cost per Saleable Unit

Every other calculation depends on getting this denominator right. It is also where most comparisons quietly go wrong.

From Unit Price to Landed Cost

Landed cost is unit price plus freight, destination charges, duty, clearance, insurance and inland delivery, divided by the units that arrive in saleable condition. That last qualifier matters, since a 3% defect rate raises effective landed cost by more than most negotiated discounts lower it.

Freight is the line buyers estimate most loosely. Container rates move constantly, and an index such as Drewry's World Container Index gives a weekly reference point for the major lanes so your model uses a market figure rather than last year's invoice.

Carton Geometry Decides More Than You Think

Freight is sold by volume, so how many units fit a container is a pricing variable you control. A carton redesign that improves container utilisation by 15% often beats a negotiated discount of the same size, and it applies to every shipment afterwards rather than to one order.

Ask for carton dimensions and units per carton in the first quote round from both routes. Then calculate units per container yourself rather than accepting a per-unit freight estimate, since the two frequently differ by a margin large enough to reverse a decision.

Currency belongs in the denominator as well. Where one route quotes in dollars and the other in yuan, a movement between order and payment lands entirely on you, and on a six-figure order a two percent swing outweighs most negotiated discounts. Note the quoting currency for each line and decide deliberately whether you are carrying that exposure.

Common Mistake to Avoid:Comparing a factory quote against an intermediary quote without equalising scope is the error that makes almost every one of these analyses wrong. The factory number covers goods leaving a port. The intermediary number often covers goods arriving at your warehouse with quality checked along the way. Subtracting one from the other produces a saving that does not exist. Build both routes to the same endpoint with the same services included, and the gap usually narrows to something like a third of what the raw quotes suggested.

The Costs Buyers Leave Out

Four lines account for most of the difference between a model that predicts reality and one that flatters a decision already made.

Minimum Order Overhang and Cash

Overhang is the gap between what you must buy and what you can sell within a sensible period. If a factory minimum is 3,000 and you sell 1,800 a year, you are financing 1,200 units of stock for roughly eight months, plus storage, plus the risk that the product stops selling. Multiply the surplus by unit cost and treat it as a cost of the route, not as inventory.

Cash sits alongside it. A direct order commonly ties up money from deposit through production, transit and sell-through, and that period is often two to three times longer than buying smaller quantities from a stockholding intermediary. Apply your real cost of capital, or the return you would have made deploying that cash elsewhere.

Inspection, Defects and Your Own Hours

Inspection economics cut sharply by volume, since sampling standards such as ISO 2859-1 size the sample against the lot and the sample grows far more slowly than the batch. A fixed inspection fee is trivial per unit on 5,000 pieces and prohibitive on 300, which is exactly why small direct orders so often ship unchecked and why defect losses cluster there.

Time is the line nobody writes down. Supplier search, sample rounds, specification writing, chasing production, arranging freight and resolving problems all consume hours that would otherwise go somewhere productive. Count them honestly for the first order and again for a repeat order, because the first is always far heavier and buyers tend to model the second.

Rework and returns deserve their own line rather than being folded into defects. A batch that is saleable after sorting costs you labour and delay rather than the goods themselves, and marketplace returns carry fees plus the reputational cost of poor reviews. Both are real, both are route-dependent, and neither appears on any quotation.

Expert Tip:Track hours on your first direct order in a simple log rather than estimating afterwards. Date, task, minutes. The total is reliably two to three times what people guess, and the shape of it is instructive: most of the hours go into clarification rather than negotiation. That finding usually changes what buyers want from a supplier relationship more than the total figure does.

Finding Your Break-Even

With the nine lines populated, the question becomes arithmetic rather than opinion. Two variables move the answer more than anything else.

Volume per Product and Number of Products

Annual volume per product determines whether fixed costs amortise. Sampling, setup, inspection and your search time are largely fixed per product, so they fall per unit as volume rises and stay stubbornly high on a product selling a few hundred a year.

The number of products determines whether your time survives. Direct management of two products is demanding and possible. Direct management of fifteen across four categories is a full-time job, and at that point the comparison stops being about price and becomes about whether you are running a purchasing department by accident.

Model a range rather than a single figure. Run the comparison at your realistic sales rate, then again at 70% of it, since the pessimistic case is where minimum order overhang shows its true weight. A route that survives the lower case is a decision you can commit to. One that only works at the optimistic number is a bet dressed as a calculation.

What Moves the Break-Even Point

Direct buying wins earlier when the product is stable, high volume, low complexity and bought repeatedly from the same supplier. It wins later, or never, when products change seasonally, when volumes are modest, when categories are varied, or when specification is complex enough that oversight needs technical judgement.

Duty treatment can shift the picture too, since rates follow the HS classification of the goods rather than the route by which you bought them. Confirm the code before modelling, because an assumed rate that turns out wrong moves landed cost by more than most sourcing decisions do.

Expert Tip:Run the model on a product you already buy before applying it to a new one. You have real numbers for sales rate, defect experience, freight and hours, so the output is testable against what actually happened. Buyers who build the model on a hypothetical product get a spreadsheet that agrees with whatever they hoped. Building it on history first produces something you can trust for the next decision.

Where NewBuyingAgent Sits in the Model

For many buyers, the real difference between sourcing direct and working with a buying partner is not simply the product price. It is the total cost of finding suppliers, comparing offers, communicating across factories, following up on orders, and coordinating multiple products.

NewBuyingAgent uses its network of factories across China to source products based on the buyer’s requirements, rather than limiting the search to a single supplier. This can help buyers access more competitive factory pricing and reduce purchasing costs by around 5%–10%, depending on the product and order.

More importantly, NewBuyingAgent takes over day-to-day supplier communication and purchasing coordination. This is particularly useful for buyers sourcing multiple product categories, where managing several factories directly can quickly become a significant time commitment.

The goal is not simply to add another layer between the buyer and the factory. It is to make the overall sourcing process more efficient—so buyers can spend less time managing China suppliers and more time growing their business.

Frequently Asked Questions

How much cheaper is buying direct from a Chinese factory?

At the unit price level the gap often looks like 10% to 25%. Once freight, inspection, minimum order overhang, cash and your own time are counted, the realised difference is usually far smaller and can be negative on low-volume products. Build the comparison to the same endpoint before treating any headline figure as a saving.

What is the biggest hidden cost of going direct?

Minimum order overhang on most low and mid-volume products, because it scales with the very discount that attracted you. Stock bought beyond your sell-through absorbs cash and storage and carries obsolescence risk. For buyers running many products, their own hours usually overtake it as the largest line.

At what point does direct sourcing start to pay?

It depends on volume per product rather than total spend. Products selling in the thousands per year with stable specifications amortise the fixed costs and generally justify direct buying. Products selling in the hundreds rarely do, however attractive the unit price looks, because the fixed lines never spread far enough.

Can I go direct on some products and not others?

Yes, and it is usually the right answer. Run your highest-volume stable products direct where the arithmetic works, and use an intermediary for the long tail of smaller and seasonal items. Applying one model to a whole range is what produces the awkward middle, where neither route is being used well.

How often should I rerun this comparison?

Annually for products you already run, and whenever volume changes materially or freight rates move sharply. Break-even points shift as sales grow, so a product that belonged with an intermediary two years ago may now justify direct buying. The model takes minutes to update once the structure exists.

Conclusion

Direct buying is neither cheaper nor dearer as a general rule. It is cheaper above a volume threshold that depends on your product, your sell-through and what your own hours are worth. Populate the nine lines, build both routes to the same endpoint, and the decision usually answers itself, product by product rather than as a policy.For products that fall on the wrong side of that threshold, NewBuyingAgent handles factory selection, quality control and delivery from China.

Sources

1. International Chamber of Commerce — Incoterms rules, the international standard for allocating cost and risk between buyer and seller —https://iccwbo.org/business-solutions/incoterms-rules/— accessed 6 August 2026

2. Drewry Supply Chain Advisors — World Container Index, weekly composite spot rate per 40ft container across major east-west trade lanes —https://www.drewry.co.uk/supply-chain-advisors/supply-chain-expertise/world-container-index-assessed-by-drewry— accessed 6 August 2026

3. International Organization for Standardization — ISO 2859-1:2026, Sampling procedures for inspection by attributes, Part 1, in which sample size is determined by lot size —https://www.iso.org/standard/85464.html— accessed 6 August 2026