Five projects, five different problems, one shared cause: nobody measured the process until something went wrong. These are the case files behind our production and quality pages. Every figure below comes from the project record, not from a summary written afterwards.
Five Projects, Five Different Problems
| Project | The problem | What changed | Measured outcome |
|---|---|---|---|
| Scandinavian brand | Defect rate stuck at 11% across three suppliers | Defect taxonomy agreed first, then station-level sign-off | 11% to 1.2%, without switching factories |
| Importer, trading company | A "factory" that was reselling | Verification before payment, not after | $200,000 lost on the earlier order |
| Mid-size importer | FOB too high but no room to change supplier | Cost conversation instead of a supplier change | FOB down 12%, same factory, same quality, same delivery |
| Northline, front-opening carry-on | One-page brief, fixed launch date | Tooling and sampling run in parallel, not in sequence | 20,000 units in 14 weeks |
| Smart luggage line | A battery-powered case with certification exposure | Mold flow simulation before tooling; UN38.3 and IATA rules designed in | Line released to mass production |
Read the table by column, not by row. The third column is the only one you can copy, and in four of the five projects it is a measurement change rather than a supplier change.
What the Five Have in Common
In four of the five, the fix was not a new factory. It was making a number visible that nobody had agreed on: which defect counts as major, which station owns it, and when it gets signed off.
The fifth is the exception, and it is the expensive one. A trading company pretending to be a factory does not fail on quality — it fails on disclosure. By the time the defect rate shows it, the money is already gone.
Where to Start, Depending on Your Problem
If your defect rate is high and you do not know which station causes it, start with the defect taxonomy, not with a new supplier. You cannot negotiate a rate you cannot name.
If your unit price is high and your specification is fixed, the saving is usually in material utilisation and packaging, not in labour. That is a conversation, not a re-tender.
If you have not yet placed an order, the cheapest hour you will ever spend is verifying that the entity you are paying is the entity that owns the machines.
How We Measure It
Every project above runs on the same three records: a first-article inspection before the line is released, a station sign-off sheet that names the person and the time, and a pre-shipment inspection against the agreed defect levels. The defect rate in the first case is a before-and-after of the same three records, not a new process.
That is also why we can quote a range instead of a single number. Where the data does not exist, we say so rather than estimate.
What to Send Us
If you want this applied to your own project, send three things: your current defect rate if you have one, the last inspection report you received, and the specification you are working from. We will tell you which of the five patterns you are in before quoting anything.
If you have no inspection report yet, say that instead — it changes the starting point, not the answer.
Read the Files and the Method
- Case Study #1 — defect rate from 11% to 1.2%
- Case Study #2 — the $200,000 lesson on factory verification
- Case Study #3 — FOB down 12% without changing suppliers
- Northline — 20,000 units in 14 weeks
- Smart luggage — sketch to mass production
The method behind them: quality control in luggage manufacturing, how to calculate your total cost, and the evidence behind our factory claims.