Test shipment: why, what volume and what to measure

20.08.2026·13 min read read·Delivery Pro China

A test shipment is a real delivery handled by a new logistics provider alongside your current one, so you can compare results rather than promises. A small shipment is usually enough: it goes through every stage of the process and shows you the final cost at your warehouse, the actual timeline for each stage and how the provider responds to the first unexpected issue.

Test format

in parallel with the current contractor

Minimum volume

50 kg, better value to consolidate from 150 kg

Main metric

total at your warehouse per unit of usable goods

Quote validity

7 days

What we record before the test

price, timing by stage, scope of services

Why run a test before moving all shipments

An established importer already has a working arrangement. Even if it is inconvenient and more expensive than you would like, it is predictable: you know the timing, you know whom to call, and you know where problems usually arise. Moving the entire volume to a new contractor on the strength of a commercial proposal trades predictability for the unknown, and does so across the whole purchasing programme at once.

A test splits that decision in two. First you get the facts on one batch, then you decide on volume. The cost of a mistake in the first case is limited to a single delivery; in the second it is a quarter of sales.

There is a second reason, unrelated to risk. A test gives you comparison data that neither you nor the operator has at present. Before the test the conversation is about rates and promises; afterwards it is about specific lines in a calculation and specific dates. This is the only way to test the full shipment cost method on your own figures.

Finally, a test checks not the logistics but the fit. How the operator asks questions, how it reacts to a change in the order, what it does when there is a discrepancy, how easy your buyer finds it to work with them. None of this is measured in a per-kilogram rate, yet it is exactly what determines how much of your team's time shipments will take.

Why the test runs in parallel with the current contractor rather than instead of it

Replacing your contractor for the duration of the test deprives you of a reference point. If the current arrangement is stopped, there is nothing to compare the new figures with – all that remains is a subjective feeling that things got better or worse. A parallel test gives you two deliveries in the same period and under the same market conditions.

The second motive is practical. While the test runs you are not left without goods if something goes wrong. The absence of deadline pressure also changes the quality of your observation: you watch the process instead of fighting a fire.

The third motive concerns suppliers. In a parallel test you are not moving all your purchasing from a supplier to a new operator; you hand over part of an order or a single SKU. The relationship with the supplier does not change, and they do not read it as a change in the terms of business.

What volume is sufficient for a test

A test shipment should be large enough to go through every stage of the standard process without being handled as an exception. We accept shipments from 50 kg, which is usually sufficient. The test is about the process, and its core stages are the same for small and large shipments: collection from the supplier, goods receipt, inspection, packing, freight, documentation and last-mile delivery.

A shipment of that size gives you every line in the quote and a date for every stage. It will not show the effects of scale, such as how the provider handles a full vehicle or palletised freight. Price also needs context: the cost per kilogram on a minimum-size shipment is noticeably higher than on a shipment of 150 kg or more, because fixed costs – collection, goods receipt, packing and last-mile delivery – are spread over fewer kilograms. The test tells you whether the process works, not what annual-volume pricing will look like.

A reasonable benchmark for choosing the volume is below.

• The batch passes through every stage without manual exceptions or shortcuts
• It contains more than one SKU: otherwise item reconciliation is not tested
• Preferably two suppliers: this tests consolidation and synchronisation
• The goods are typical for you in density and fragility, not the most convenient ones
• The value of the batch is such that losing it does not ruin your month
• The timing is not tied to an advertising launch date or a commitment to a customer

The second most common question is how many batches are needed. One batch shows the price and the timing. Two batches show repeatability, and for regular purchasing that matters more than price. If your schedule allows, it is better to plan two in a row and see whether they are calculated under the same rules.

What to measure during the test

Six groups of parameters are measured. Five of them are visible straight away; the sixth only if something did not go to plan, and it is the most informative.

Final cost at your own warehouse

Not the per-kilogram rate but the sum of all blocks up to the moment of unloading. It is calculated per unit of usable goods: the total is divided by the number of units you can sell, not by the number ordered. It is this figure that is compared with your current arrangement, calculated under the same rules.

Actual timing by stage

A single overall lead time says little. What you need are dates by stage: payment to the supplier, dispatch from the supplier, goods receipt at the warehouse, readiness for dispatch, departure of the shipment, arrival in the city, unloading at your premises. That shows exactly where time is lost and who influences it.

Condition of the cargo

Number of pieces, integrity of the packaging, defects and shortfall in the end. Separately: whether what was recorded at goods receipt in China matches what you received at your own premises. A difference between these two points points to the carriage; a match points to the supplier.

Completeness of documents

Whether the set your accounting department needs arrived, and whether it arrived on time. This is checked on the actual set, not on a list from a presentation: invoice, packing list, transport documents and, under official import, the full declaration set.

Quality of communication

How many times you had to ask about status yourself. Whether updates arrived before you requested them. How many people you had to keep in the correspondence and whether you had to connect the participants in the chain yourself. This is a direct measure of the load on your buyer.

The first non-standard situation

The most valuable outcome of a test. The supplier shipped the wrong thing, part of the goods was late, the cargo was delayed, the rate changed – what matters is what happened next. Whether you were warned in advance or found out after the event, whether you were offered options or informed of a decision, whether the price changed unilaterally.

What a test will not show

A single batch has a limit to how much it can tell you. Below is what a test does not check, and it is more honest to know this in advance than to draw conclusions from one delivery that it cannot support.

• Behaviour at volume: discounts, filling a vehicle, handling palletised cargo
• Resilience to seasonal peaks: before Chinese New Year the chain behaves differently
• Stability of timing: one batch gives a point, not a range
• Behaviour when market rates change, if they did not change during the test
• Handling of rare categories, if the test goods did not belong to them
• What happens in a major failure: a test usually runs without serious incidents
• Long-term handling of claims against suppliers, which takes weeks
• The quality of work with official import, if the test ran under the cargo scheme

From this follows a simple conclusion: a test answers the question of whether to increase the new operator's share, not the question of whether to move everything at once. The second question is settled at volume.

How to record the terms before the test

If the starting terms are not written down, there will be nothing to compare against once the test is over: memory adjusts to fit the result. Recording them takes half an hour and is done once, before the order.

1. Describe the cargo: items, quantities, suppliers and their cities, weight and dimensions of each piece
2. Write down the current figures: what this same batch would cost under your existing arrangement, by the same blocks
3. Write down the current timing by stage, not one figure from order to receipt
4. Obtain the new operator's calculation as a breakdown by block, with a date and a validity period
5. Record the scope of services: what is included, what is charged separately, what is not available at all
6. Record the chargeable weight rule and what happens if the dimensions change
7. Record the rules for price changes: under what conditions the invoice may differ from the calculation
8. Agree the inspection brief: what to open, count, measure and photograph
9. Agree what to do by default if there is a discrepancy: wait for a top-up delivery or ship the good units
10. Appoint who on your team runs the test and where the documents and correspondence are kept

Rates change, so compare quotes that are valid on the same date. Compare the recorded figure with your current provider's quote for the same period, not with last year's invoice.

Figures to agree before sending the test shipment

A quote with no specific amounts gives you nothing to compare. The figures below are the ones we provide before a test shipment, as at 20 August 2026. Ask every provider for the same information, including your current one.

These ranges are the baseline for assessing the actual result. If the international leg takes longer, ask which stage caused the delay instead of debating the headline lead time. Record the timing of each stage so any delay can be traced. Mandatory cover for the international leg is provided by the carrier. This cover does not include cargo damage; liability for damage is determined by the contract of carriage and any applicable convention.

Twelve points: what to compare in an operator after the test

Go through the list immediately after unloading, while the correspondence and documents are to hand. Each point is closed by a fact, not an impression.

1. The final cost of the batch at your warehouse, calculated per unit of usable goods, against the same figure under your current arrangement
2. Whether the final invoice matched the original calculation line by line, and exactly what diverged
3. Actual timing for each stage against what was promised, with a note of where time was lost
4. The chargeable weight on the invoice: by actual or volumetric weight, and whether the factor was stated in advance
5. The actual scope of services: what you had to buy in addition to the original scope
6. Goods receipt in China: whether you received a photo report, and whether the quantity can be counted and the SKUs read from it
7. Discrepancies found before dispatch, and what was done about them before the international carriage
8. Defects and shortfall at your own goods receipt: how many, and whether this matches the receipt in Guangzhou
9. Condition of the packaging after the carriage and the adequacy of the cargo preparation
10. Completeness and timeliness of the documents for your accounting department
11. Communication: how many times you asked about status yourself and whom you had to connect with whom
12. The first non-standard situation: when you were warned, what options you were given and who paid for the consequences

A useful habit is to fill in the same list for your current contractor on their next batch. Comparing two completed lists is more informative than any argument about rates.

What the transition looks like after a successful test

A successful test is not an instruction to move everything. The sensible approach is to increase the share in stages, starting with the destinations and categories where the new arrangement produced a measurable advantage.

How this works with us is described on the page about moving from your current contractor. Individual blocks of the process – payment to the supplier, the warehouse in Guangzhou, inspection of goods and consolidation – can be tested separately if you do not need the whole cycle at once.

Common mistakes in a test

The temptation to make the test as convenient as possible deserves a separate mention. If you hand over the densest goods from your most disciplined supplier to the most convenient city, anyone will pass the test. A test is useful in exactly the measure that it resembles your normal purchasing.

If you want to start, put together a description of your next batch and request a breakdown in the calculator. Questions about terms and the minimum volume are collected in the questions section.

FAQ