10. Stay, coexist, or migrate
Selenium WebDriver · 48 pages source format
Enterprise reality: Selenium is not wrong. Decide with flake data, speed, hiring, and cost — not Twitter hype.
What you'll learn
- Migration criteria
- Hybrid coexistence
- Playwright/Cypress comparison
- Decision memo
Honest comparison
Selenium: language flexibility, Grid maturity, huge legacy. Playwright/Cypress: auto-wait, tracing, faster authoring. Tradeoffs, not winners.
Do this now
Comparison table in MIGRATION.md: 6 dimensions (speed, flakes, hiring, CI cost, learning curve, legacy).
Migration criteria
Migrate when: flake rate unsustainable, CI time blocks team, hiring cannot find Selenium skill, no ROI on Grid maintenance.
Do this now
Write criteria list — 5 triggers to migrate, 5 reasons to stay.
Hybrid pattern
New flows in Playwright, legacy Selenium until ROI proves rewrite. API layer shared.
Do this now
Sketch hybrid architecture diagram for a fictional 500-test Selenium shop.
Pilot one flow
Port one stable flow to Playwright or Cypress. Compare CI time, flake rate, lines of code.
Do this now
Port login + inventory OR document pilot plan if time-boxed.