← Engineering notebook
Evaluations & production

How to run customer evaluations properly

A successful evaluation proves where the solution stops working, not just where it works.

Every sample lies - not because anyone is misleading, but because evaluation samples rarely represent reality.

Common issues with sample sets

  • Sample size too small.
  • Missing normal production variation.
  • Supplier differences not represented.
  • Only a handful of defect examples.

My evaluation process

  1. Prove feasibility - can it work at all?
  2. Gather variation - get samples across suppliers, lines, and time.
  3. Stress-test assumptions - actively try to break it.
  4. Deploy carefully - ramp slowly and watch for surprises.

The lesson

A successful evaluation proves where the solution stops working, not just where it works.