ꦲꦏ꧀ꦱꦫ
Back to stories
<QA/>
QATestingAutomationQA

Test Automation Pyramid: Unit, API, and UI

PublishedUpdated AKSR
Share by

Test Automation Pyramid: Unit, API, and UI

The test automation pyramid is a model for balancing tests: many fast unit tests at the base, fewer API tests in the middle, and fewer slow UI tests at the top. This post explains the pyramid and why it helps.

The pyramid

  • Base (wide): Many unit tests. Fast, cheap to run, focused on one unit (function, class). Usually written by developers; run on every commit.
  • Middle: Fewer API/integration tests. Slower than unit, but faster than UI; test contracts and business logic. QA and developers; run in CI.
  • Top (narrow): Few UI/E2E tests. Slow, brittle, expensive to maintain. Cover critical user journeys; run on every build or before release.

Why this balance

  • Speed: Unit and API tests run in seconds or minutes; UI tests can take much longer. More at the base = faster feedback.
  • Stability: Unit and API tests are less flaky than UI (no browser, no DOM). Fewer UI tests = fewer flaky failures.
  • Maintenance: UI tests break when layout or flow changes. Fewer UI tests = less maintenance.
  • Coverage: Unit tests cover logic; API tests cover integration and contracts; UI tests cover end-to-end flow. Together they give broad coverage with reasonable cost.

In practice

  • Prioritize unit tests (developers) and API tests (QA + dev); add UI tests only for critical paths.
  • Run unit and API tests on every commit or push; run full UI suite on schedule or before release.
  • When in doubt, add an API test rather than another UI test.

Summary

  • Pyramid = many unit, fewer API, few UI tests; fast and stable at the base, slow and costly at the top.
  • This balance gives fast feedback, less flakiness, and lower maintenance.
  • Invest in unit and API automation first; use UI automation for critical journeys only.