Types of testing
Functional vs non-functional, smoke, sanity, regression, integration, UAT and more — with when to use each.
Testing types answer two questions: what are we checking, and at which stage?
By level
- Unit testing — individual functions or classes, usually by developers.
- Integration testing — modules working together (for example frontend + API + database).
- System testing — the complete application against requirements.
- User acceptance testing (UAT) — real users or the client confirm it meets their needs.
Common types
- Smoke testing — a quick check that a new build's main features work, before deeper testing.
- Sanity testing — a narrow check that a specific fix or change works.
- Regression testing — re-running existing tests to make sure new changes didn't break old features. Prime candidate for automation.
- Exploratory testing — unscripted testing guided by the tester's experience and curiosity.
Non-functional
Performance (speed under load), security, usability, compatibility (browsers, devices) and accessibility.
Black-box testing checks behaviour without looking at code; white-box testing uses knowledge of the code paths.
Example
Practise on any app you useA new build arrives with a fix for the cart-discount bug:
1. Smoke — can users log in, search, add to cart, pay? (15 min) 2. Sanity — does the cart discount now update correctly? (10 min) 3. Regression— coupons, quantities, shipping, taxes, payments still correct across the checkout? (2–3 hrs, mostly automated) 4. Release candidate → UAT by the client's operations team
What to notice
• Smoke decides whether the build is worth testing at all.
• Sanity confirms the specific fix.
• Regression protects everything around the fix — fixes often break neighbouring features.
Apps, IDs and data are fictional examples written for practice.
Key points
- Levels: unit → integration → system → UAT.
- Smoke = build is stable; sanity = specific fix works.
- Regression protects existing features; automate it.
- Remember non-functional testing too.
Quick check
Which testing re-checks old features after a change?
Regression testing confirms that new changes haven't broken features that already worked.
Practice task
For a food-delivery app release that changes the address screen, list what you'd include in smoke, sanity and regression testing.
Interview corner
Smoke vs sanity testing?
Smoke is a broad, shallow check of a new build's critical paths. Sanity is a narrow, focused check that a specific change or fix works.
Which tests would you automate first?
Stable, repetitive, high-value ones — regression suites and smoke checks on critical flows. Avoid automating features that change constantly.
Original lesson written by the InfoparkDaily team · © 2026 InfoparkDaily · Free to learn — no paid course or subscription.