← All breakdowns

The sample was fine. The bulk order was not.

Nothing went wrong deliberately. The sample was made under one set of conditions and the order under another, and nobody signed off on the difference.

Almost no factory sets out to send you something worse than the sample. What happens is more ordinary: the sample is made under one set of conditions, the bulk order under another, and nobody changes anything deliberately.

Three things change, and all three are invisible in a photograph

Three things that change between sample and bulk Material grade, tooling condition and production line. Each is usually correct at sample stage and degrades at volume. What you approved, and what arrives Sample Bulk run Material grade and virgin content Stated grade, all virgin Regrind blended in Tooling the mould that made it New, or hand-finished Worn mould, flash trimmed Line who actually ran it Sample team, slow pace Volume line, steps cut

Each line is a decision someone makes for a reason that sounds sensible in the moment. Regrind is free material and reduces cost. An older mould is already set up. The volume line runs faster with two inspection steps instead of four. Nobody is lying to you; the order simply gets optimised by people who are not thinking about the standard you approved.

Why it is hard to prove afterwards

All three changes leave traces that are expensive or impossible to read after the fact. Material grade needs a lab. Mould condition disappears once the parts are packed. Line changes are not documented anywhere you can reach.

So the work is not in detecting it later. It is in closing the three doors before production starts, and in looking at the right moment.

Four things that actually hold

  1. A signed golden sample, held by both sides. Not a photo — a physical unit, sealed, dated, signed, one copy each. Every dispute about "this is within tolerance" becomes a comparison against an object rather than a memory.
  2. Material named at grade level, with batch documentation. A contract that says "PP" is not a specification. It needs the grade, the supplier of the resin, and the factory's agreement to hand over the material batch certificate with the shipment.
  3. First-article approval in writing. The first units off the volume line, not off the sample bench, approved by you before the run continues. This is the single cheapest control in the whole process and it is the one most often skipped.
  4. An in-line inspection, not only a pre-shipment one. This is the point most buyers miss, and the next figure is why.
Inspection timing across a production run First-article approval at zero, in-line inspection around thirty percent where changes are still possible, pre-shipment at eighty percent where only sorting is possible. The same defect costs different amounts depending on when you find it 0% first article 30% in-line: fixable 80% pre-shipment 100% shipped Before 30%: the factory can still change the material, the setting or the mould. After 80%: your only remaining options are sorting, rework, or shipping it anyway. This is why first orders get an in-line inspection and repeat orders often do not.
Pre-shipment inspection is not a quality control. It is a decision point.

A pre-shipment inspection happens when the goods are made. If it finds a problem, the factory can sort, rework, or you can cancel — and cancelling at that stage means you have paid for goods you cannot use and a container you cannot fill. An in-line inspection at roughly a third of the run happens while the answer is still "adjust the setting".

What none of this prevents

A factory that substitutes material in a way no document catches. A production manager who decides your order is small enough to fit around a bigger one. A mould that fails halfway through the run, forcing a repair whose effect on dimensions nobody measures.

Controls move the odds; they do not close the gap. The honest version of what an inspection buys you is a documented snapshot at a known time — which is far more than nothing, and considerably less than a guarantee.

The reason to spend on this at the first order rather than the fifth is asymmetry. On a first order you have no working relationship, no history of how this factory behaves under pressure, and no idea which of the three changes it is most likely to make. After two or three orders you will know, and the controls can thin out.

An in-line inspection costs a fraction of a container

The 32-point audit includes material and tooling verification, and a pre-shipment inspection is $199.

Send a supplier to check