Code Coverage and Quality Metrics

What Code Coverage Measures

Code coverage measures what percentage of your source code is executed when your tests run.

Four types of coverage:

Line coverage — Which lines of code were executed?

Branch coverage — Which branches (if/else, ternary) were taken?

Function coverage — Which functions were called?

Statement coverage — Which statements executed?

Branch coverage is the most meaningful — it catches logic that is only wrong in certain conditions.

Setting Up Coverage in Vitest

// vitest.config.ts
export default defineConfig({
  test: {
    coverage: {
      provider: 'v8',
      reporter: ['text', 'html', 'lcov'],
      exclude: [
        'node_modules/**',
        'src/test/**',
        '**/*.config.ts',
        '**/*.d.ts',
      ],
      thresholds: {
        lines: 80,
        branches: 70,
        functions: 80,
        statements: 80,
      },
    },
  },
});

npx vitest --coverage

This generates an HTML report at coverage/index.html. Red lines were never executed; yellow branches were partially covered.

The Coverage Trap

100% coverage does not mean zero bugs.

Coverage measures execution, not correctness. A test that calls a function and asserts nothing will count as coverage.

// This test achieves 100% line coverage of add()
it('covers add', () => {
  add(2, 3); // Executes every line
  // No assertion — verifies nothing
});

// This is the same coverage but actually tests behavior:
it('adds two numbers', () => {
  expect(add(2, 3)).toBe(5);
  expect(add(-1, 1)).toBe(0);
});

Coverage is a floor, not a ceiling. Use it to find gaps, not as a quality guarantee.

Practical Coverage Targets

Pursue 100% coverage in your most critical, high-risk code. Accept lower coverage in UI and exploratory code.

Mutation Testing: Coverage Done Right

Mutation testing answers: "If I introduce a bug, does a test fail?"

It works by automatically introducing small bugs (mutations) — changing > to >=, removing a condition, negating a boolean — and running your tests. If tests still pass with a bug present, they're not actually catching that bug.

# Stryker is the leading mutation testing tool
npm install -D @stryker-mutator/core @stryker-mutator/vitest-runner
npx stryker run

Mutation score is a better quality metric than line coverage.

Key Takeaways

• Branch coverage is more meaningful than line coverage — it catches conditional logic bugs
• Set coverage thresholds that fail CI if coverage drops — this creates accountability
• 100% coverage does not mean zero bugs — coverage measures execution, not correctness
• Focus high coverage (95%+) on critical paths: authentication, payments, data mutations
• Mutation testing is the gold standard — it verifies that your tests would actually catch bugs