Testing

The Testing Pyramid

The testing pyramid is a concept that helps guide how many tests of each type you should write:

        /\
       /  \      ← UI Tests (Few, Slow, High-level)
      /────\
     /      \    ← Integration Tests (Some, Medium speed)
    /────────\
   /          \  ← Unit Tests (Many, Fast, Low-level)
  /────────────\

Recommended distribution:

• 70% Unit tests - Fast, isolated, test individual functions/methods
• 20% Integration tests - Test components working together
• 10% UI tests - Test critical user journeys end-to-end

Why this ratio?

• Unit tests are fast and catch most bugs early
• Too many UI tests = slow test suite and high maintenance
• UI tests focus on critical paths, not every edge case

Comparison of test types

Aspect	Unit Tests	Integration Tests	UI Tests
Scope	Single function/method	Multiple components	Entire application
Speed	⚡⚡⚡ Very fast	⚡⚡ Fast	🐌 Slow
Reliability	✅ Very stable	✅ Stable	⚠️ Can be flaky
Maintenance	✅ Low	🟡 Medium	⚠️ High
Debugging	✅ Easy	🟡 Medium	⚠️ Difficult
Coverage	Detailed	Moderate	High-level
When to use	Always, many	For integrations	Critical paths only
Example	add(2, 3) === 5	API + DB test	Full checkout flow

Further information

Link: UI Tests