How to use it
Paste this into a spreadsheet, with one row per test case. Add a new result column for each version so you can compare them.
The template
Case ID Input Correct answer Version 1 output Pass or fail Error type Severity (low/med/high) Notes SUMMARY Version: Date: Cases run: Passed: High-severity failures: Most common error type: Decision (ship / fix / stop):
A filled-in example
Case ID: 014 Input: "Charged twice for March, please fix" Correct answer: billing, not urgent Version 1 output: billing, urgent Pass or fail: fail Error type: wrong urgency Severity: low Notes: over-marks billing as urgent. Tighten the urgency rule.