Metrics for Testing Print

  • softwaretestingqa, software, troubleshooting, guide, howto, solution, zillionkinghost, hosting
  • 0

Measuring the work usefully.

WHAT TO MEASURE

Defects escaping to production Time from defect introduction to detection Automated test duration and reliability Coverage of critical areas Time to verify a fix

WHY ESCAPED DEFECTS MATTER MOST

They are the outcome testing exists to prevent.

WHAT NOT TO MEASURE AS A TARGET

Number of defects found Number of tests written Coverage percentage Tests executed per day

WHY THOSE CORRUPT

They measure activity, and each can be increased without improving quality.

WHAT COUNTING DEFECTS FOUND ENCOURAGES

Reporting trivia, and adversarial relationships.

WHAT COUNTING TESTS ENCOURAGES

Many small tests verifying little.

WHAT TO TRACK ABOUT DEFECTS

Where they were introduced Where they were found Their cause category

WHAT THAT REVEALS

Which stage is leaking, and why.

WHAT TO REPORT TO STAKEHOLDERS

Risk: what is tested, what is not, and what remains uncertain.

WHY THAT FRAMING

It supports the decision they actually face, which is whether to release.

WHAT TO NEVER REPORT

A percentage implying completeness.

WHAT TO REVIEW

Whether each metric changes any decision.


Was this answer helpful?
Back

Are you happy with your experience? Leave us a review on Trustpilot.


Trustpilot