← Engineering notebook
Evaluations & production

Every sample lies

Customers send their best and worst parts. The dangerous gap is everything in between.

This became one of my favorite engineering sayings. Not because customers are misleading, but because evaluation samples rarely represent reality.

Evaluation samples cover a narrow range; production is much wider
Samples cluster in a narrow range. Real production is far wider - and its tails are what cross the defect threshold.

The typical situation

A customer sends:

  • Their best samples
  • Their worst samples

What is almost always missing:

  • Normal production variation - the boring middle that makes up 99% of real output.

The result

The evaluation looks excellent. Then production performs differently, because production is mostly the variation that was never in the sample box.

My current questions

Before trusting any sample set, I ask:

  • How many suppliers?
  • How many production lines?
  • How many years of process history?
  • What recently changed?