Testing Strategy Models

Model Comparison

Model	Structure	Best For
Pyramid	Unit 70% > Integration 20% > E2E 10%	Monoliths, logic-heavy
Trophy (Dodds)	Static > Integration (largest) > Unit > E2E	Modern SPAs
Honeycomb (Spotify)	Contract-centric cells	Microservices
Diamond	Balanced unit/integration	Domain services

Testing Trophy (Recommended for SPAs)

         E2E (minimal)
        /------------\
       / Integration  \   <-- Largest portion
      /----------------\
     /   Unit Tests     \
    /--------------------\
   /   Static Analysis    \  <-- Foundation
  /________________________\

Philosophy: "The more your tests resemble how software is used, the more confidence they give you." - Kent C. Dodds

Key Principles:

Test behavior, not implementation
Minimize mocking
Prioritize integration tests
Use accessible queries (getByRole first)

Testing Honeycomb (Microservices)

Contract testing at center, interconnected cells for:

Unit tests (implementation details)
Integration tests (service boundaries)
Contract tests (API agreements)
E2E tests (critical paths only)

Ratios by Context

Context	Unit	Integration	E2E
Classic Pyramid	70%	20%	10%
Testing Trophy	30%	50%	10%
API-heavy	75%	15%	10%
Microservices	40%	40%	20%

Priority Matrix

Priority	Category	Examples
P0	Core flows	Signup, login, checkout, payment
P1	Major features	Search, CRUD, navigation
P2	Secondary	Filters, sorting, pagination
P3	Edge cases	Empty states, max limits

Coverage Targets

Area	Target
Critical paths	100%
Core features	80-90%
Overall	75-85%

Note: Coverage as diagnostic, not target. Focus on what's uncovered.

CI/CD Order

- run: npm run lint          # Gate 0: Static analysis
- run: npm run test:unit     # Gate 1: Fast fail
- run: npm run test:integration # Gate 2
- run: npm run test:e2e      # Gate 3: Pre-merge

2.1 KiB Raw Permalink Blame History