Belajar Quality Engineer - Quality Metrics & DORA

#Pendahuluan

Setelah di episode 10 kita mempelajari test environments & test ops, pada episode ini kita mempelajari quality metrics & DORA — bagaimana mengukur kualitas menggunakan DORA metrics dan quality scorecards. Metrics yang tepat membantu tim memahami dampak dari improvement yang dilakukan.

Mengapa quality metrics & DORA penting? Karena DORA metrics adalah standar industri untuk mengukur software delivery performance. Dengan memahami metrics ini, kalian bisa benchmark performa tim dan mengidentifikasi area perbaikan.

#DORA Metrics

#4 Key Metrics

text

DORA Metrics:
├── Deployment Frequency:
│   ├── Seberapa sering deploy ke production
│   ├── Elite: On-demand (multiple per day)
│   └── Target: Daily atau lebih sering
├── Lead Time for Changes:
│   ├── Waktu dari commit sampai deploy
│   ├── Elite: < 1 hour
│   └── Target: < 1 day
├── Change Failure Rate:
│   ├── Persentase deployment yang menyebabkan failure
│   ├── Elite: 0-15%
│   └── Target: < 5%
└── Time to Restore Service:
    ├── Waktu untuk restore dari failure
    ├── Elite: < 1 hour
    └── Target: < 1 day

#DORA Performance Levels

text

DORA Performance Levels:
├── Elite:
│   ├── Deployment: Multiple per day
│   ├── Lead Time: < 1 hour
│   ├── Failure Rate: 0-15%
│   └── Restore Time: < 1 hour
├── High:
│   ├── Deployment: Weekly
│   ├── Lead Time: 1 day - 1 week
│   ├── Failure Rate: 16-30%
│   └── Restore Time: < 1 day
├── Medium:
│   ├── Deployment: Monthly
│   ├── Lead Time: 1 week - 1 month
│   ├── Failure Rate: 16-30%
│   └── Restore Time: 1 day - 1 week
└── Low:
    ├── Deployment: Monthly (less)
    ├── Lead Time: 1-6 months
    ├── Failure Rate: > 30%
    └── Restore Time: > 1 week

Note

DORA metrics bukan target yang harus dicapai sekaligus. Mulai dengan baseline measurement, lalu improve secara bertahap. Yang penting adalah trend yang improving, bukan absolute value.

#Defect Metrics

#Key Defect Metrics

text

Defect Metrics:
├── Defect Density:
│   ├── Formula: Defects / KLOC
│   ├── Target: < 1 bug/KLOC
│   └── Benchmark: industry average
├── Defect Removal Efficiency:
│   ├── Formula: (Defects found pre-release / Total defects) × 100%
│   ├── Target: > 95%
│   └── Measures: testing effectiveness
├── Defect Escape Rate:
│   ├── Formula: (Defects found in production / Total defects) × 100%
│   ├── Target: < 5%
│   └── Measures: testing coverage
└── Mean Time to Detect:
    ├── Average time from defect introduction to detection
    ├── Target: < 1 day
    └── Measures: feedback loop speed

#Quality Scorecards

#Quality Scorecard Components

text

Quality Scorecard:
├── Code Quality:
│   ├── Linting score
│   ├── Complexity metrics
│   └── Code review metrics
├── Test Quality:
│   ├── Coverage metrics
│   ├── Test pass rate
│   └── Flaky test rate
├── Process Quality:
│   ├── DORA metrics
│   ├── Defect metrics
│   └── Release quality
└── User Quality:
    ├── User satisfaction
    ├── Support tickets
    └── NPS score

#Contoh Scorecard

text

Quality Scorecard - Sprint 5:
├── Code Quality: 85/100
│   ├── Linting: 95
│   ├── Complexity: 80
│   └── Review: 80
├── Test Quality: 90/100
│   ├── Coverage: 90
│   ├── Pass Rate: 95
│   └── Flaky: 85
├── Process Quality: 80/100
│   ├── Deployment Freq: 80
│   ├── Lead Time: 75
│   └── Failure Rate: 85
└── Overall: 85/100

#Practical: Metrics Dashboard

yaml

metrics_dashboard:
  dora:
    - deployment_frequency
    - lead_time
    - change_failure_rate
    - time_to_restore
  defect:
    - defect_density
    - defect_removal_efficiency
    - defect_escape_rate
    - mean_time_to_detect
  quality:
    - test_coverage
    - code_quality
    - security_score
    - performance_score

Tip

Automate metrics collection dan display di dashboard yang bisa diakses semua tim. Visibility ke metrics membantu tim memahami dampak dari perubahan yang mereka lakukan.

#Penutup

Pada episode 11 ini, kalian telah mempelajari quality metrics & DORA.

Inti yang harus dibawa pulang:

DORA metrics: deployment frequency, lead time, change failure rate, time to restore.
Defect metrics: defect density, removal efficiency, escape rate.
Quality scorecards: code, test, process, user quality.
Metrics harus automated dan visible ke seluruh tim.

Di episode 12 selanjutnya, kita akan membahas performance engineering — bagaimana mengintegrasikan performance testing ke dalam development lifecycle. Sampai jumpa di episode 12!