How to Switch from Manual Testing to Automation in 2026 (90-Day Plan)
A day-by-day 90-day roadmap for manual testers moving into automation — Playwright or Selenium, coding basics, CI/CD, portfolio projects, and interview strategy that actually gets offers.
Last updated: July 11, 2026 · 10 min read
The salary gap between manual QA and automation is now 60–120% in most markets. This 90-day plan is what testers who successfully switched in 2025–2026 actually did — no filler courses, no wasted time. Start with our Career Path Roadmap for the big picture.
Why 2026 is the right moment
- Automation demand up 34% YoY; manual QA hiring flat.
- AI test-generation tools (Copilot, Cursor) mean less boilerplate — automation is easier to learn than in 2020.
- Playwright is dominant and beginner-friendly.
- Remote-first hiring means you compete globally with a strong portfolio.
The 90-day plan
| Days | Focus | Deliverable |
|---|---|---|
| 1–14 | JavaScript/TypeScript basics (or Java if selenium-heavy market) | 50 exercism.io problems |
| 15–35 | Playwright tutorial + first 20 tests on a demo app | Public GitHub repo |
| 36–50 | POM + fixtures + CI on GitHub Actions | Working framework |
| 51–65 | API testing (Postman + REST Assured or supertest) | 10 API tests in same repo |
| 66–75 | SQL for testers + basic Docker | Query cheat sheet, one dockerized run |
| 76–85 | Interview prep — Playwright Q&A, API Q&A, AI mocks | 3 mock interview scores > 80 |
| 86–90 | Apply — 20 targeted applications with tailored resume | 3+ interview calls |
The portfolio project that gets you hired
One well-built repo beats a certificate. Build a Playwright + TypeScript framework that automates a public site (e.g. saucedemo.com) with:
- POM + fixtures + parallel run
- API layer using request context
- GitHub Actions CI running on every PR
- HTML report published as artifact
- README with architecture diagram
Reposition your resume
Don't hide manual experience — reframe it. "Manual regression on 12 modules" becomes "Owned regression strategy for 12 modules; automated 40% of cases in Playwright, cutting cycle time from 3 days to 6 hrs." Score it with our ATS checker.
Handling the "you have no automation experience" objection
"In the last 90 days I've built a Playwright framework covering 60 test cases across UI and API layers, running in GitHub Actions on every PR. Here's the repo. My manual background means I write tests that catch real user pain, not just happy-path clicks."
Continue your prep
- Start a free AI mock interview
- Software Testing Career Path Roadmap
- Playwright Tutorial for Beginners
- QA Engineer vs SDET
The 12-week manual-to-automation transition plan (2026)
Switching from manual to automation is the single most common career move we coach at SoftwareTestPilot. Done right, it takes 12 weeks and adds 30-60% to your compensation. Done wrong, testers get stuck for years watching tutorials without building anything. Here is the exact plan that works in 2026.
| Phase | Weeks | Focus | Weekly output |
|---|---|---|---|
| 1 — Language basics | 1-2 | TypeScript or Java: variables, loops, functions, async | Solve 5 CodeWars kata / week |
| 2 — First framework | 3-4 | Playwright or Selenium: install, run, locators, assertions | 10 tests against a public demo app on GitHub |
| 3 — Framework patterns | 5-6 | Page Object Model, fixtures, config, reporting | Refactor week 3-4 tests into POM in a repo |
| 4 — API + data | 7-8 | Postman or REST Assured, JSON assertions, contract tests | API-first test suite for one public API |
| 5 — CI + reporting | 9-10 | GitHub Actions, JUnit reports, sharding, video on failure | Green PR pipeline on your public repo |
| 6 — Portfolio + interviews | 11-12 | Blog post about your project, ATS resume, mock interviews | 3 job applications a week |
What to build in your public portfolio
Every hiring manager we surveyed in 2026 said a GitHub repo with a working Playwright suite is worth more than any certification. Your portfolio should include:
- A README that explains the app under test, the framework choice, and how to run the tests.
- At least 20 tests covering happy paths, negative paths, and one flaky-test recovery pattern.
- A CI workflow file that runs on every push and posts results as a PR check.
- One design decision documented as an ADR (why POM, why not, why fixtures for auth reuse).
- A short blog post — your interview will refer back to it.
The three mistakes that stall 80% of career switchers
- Tutorial paralysis. Watching 40 hours of Playwright content without writing your own test. The rule: after 30 minutes of watching, code for 90 minutes.
- Skipping fundamentals. Copy-pasting page.click() without understanding async/await produces flaky code you cannot debug. Spend the first two weeks on language basics even if it feels slow.
- Never publishing. A private laptop repo does not exist to a hiring manager. Publish weekly, warts and all.
What to negotiate when you get the offer
Manual testers switching to automation are often underpaid on the first offer because they are still slotted at their old title. Come to the negotiation with: your public portfolio, salary data from the SoftwareTestPilot salary guides at /salary, and one competing offer if you can get it. The typical delta between accept-first-offer and negotiate-with-data is 20-40%.
Before you pick a course, read the job you're aiming at: the Automation QA Engineer role guide lists the exact skills hiring managers screen for, so you can reverse-engineer your learning plan from the requirements.