Manual QA Tester interview questions and answers

10 of 10 questions shown

1. juniorTest designA password field accepts 8 to 64 characters. Which boundary-value cases do you run?

   Model answer. Test 7, 8, 9, 63, 64 and 65 characters: 8, 9, 63 and 64 are accepted, 7 and 65 are rejected with a clear message. Add an empty value and a 64-character string of multi-byte Unicode, because limits are often enforced in bytes, not characters. Also check that leading and trailing spaces are handled the same way on the client and the server.

   I can answer this
2. juniorDefectsWhat does a bug report need so a developer can act without asking you questions?

   Model answer. A title that states the symptom and place, exact steps from a clean state, expected versus actual result, the environment (build, OS, browser, account, data), how often it reproduces, and evidence such as a screenshot, video or console and network log. Keep severity (impact) separate from priority (urgency), and describe the problem rather than prescribing the fix.

   Tools: Jira

   I can answer this
3. juniorProcessExplain severity versus priority and give an example where they point in opposite directions.

   Model answer. Severity is the technical impact of a defect; priority is the order in which the business wants it fixed. A misspelt company name on the home page is low severity but high priority before a launch. A crash in a rarely used admin export is high severity but can be low priority.

   I can answer this
4. juniorProcessWhat is the difference between smoke, sanity and regression testing?

   Model answer. Smoke is a shallow, wide check that a new build is stable enough to test at all. Sanity is a narrow check that a specific fix or change works. Regression re-checks existing behaviour for unintended breakage after a change. Teams use the terms loosely, so agree definitions in the test plan.

   I can answer this
5. midTest designEquivalence partitioning and boundary analysis are not enough in some cases. Which, and what do you use instead?

   Model answer. They treat inputs one at a time, so they miss interactions and state. Use pairwise or combinatorial design for many interacting options, decision tables for business rules with several conditions, and state-transition tests when behaviour depends on history, such as an order moving from paid to refunded.

   I can answer this
6. midExploratoryHow do you run a session-based exploratory test?

   Model answer. Write a charter (mission, area, 60 to 90 minute time box), test while taking notes on what you covered, bugs found and open questions, then debrief with the team. Heuristics such as CRUD, boundaries, interruptions and tours guide coverage. The output is defects, a coverage note and new ideas for scripted or automated checks.

   I can answer this
7. midRiskYou have two days for a release that normally takes five. How do you decide what to cut?

   Model answer. Rank by risk: what changed in the diff, how critical the area is (payments, login, data loss), where defects clustered before, and what real users touch most. Run smoke, critical paths and changed areas first and defer low-risk regression. Tell stakeholders explicitly what was not tested so the release decision uses the residual risk.

   I can answer this
8. midWebCheckout works in Chrome but the Pay button does nothing in Safari on iOS. How do you narrow it down?

   Model answer. Reproduce on a real device and the same iOS version, then attach Safari Web Inspector from a Mac to read the console and network tab. Look for a JavaScript feature Safari does not support yet, a blocked third-party cookie or storage call, and an overlay or fixed element swallowing the tap. Report the exact device, OS and build and the failing request or error.

   Tools: BrowserStack

   I can answer this
9. seniorStrategyHow do you decide which manual test cases to automate and which to keep manual?

   Model answer. Automate checks that are stable, repetitive, deterministic and high value: smoke, critical-path regression and data-driven cases. Keep exploratory, usability, one-off and fast-changing UI checks manual. Compare run frequency times manual time against build and maintenance cost to see whether automation pays back.

   I can answer this
10. seniorMetricsDefect escape rate rose for two releases in a row. What do you examine?

    Model answer. Escape rate is production defects divided by all defects found. Classify each escape by cause: missing requirement, untested path, environment difference or unintended regression, and check what was cut when time was short. Fix the process gap, for example a review step or a regression case, instead of blaming individuals.

    Tools: Jira, TestRail

    I can answer this

No question matches those filters. Clear the search or choose All.