Manual testing interviews check three things: whether you know the testing basics (STLC, test design techniques, the defect life cycle), whether you can apply them to a real screen such as a login page, and how you handle disagreement with developers. Tosca rounds add modules, TestCase Design and recovery. Below are 28 questions across both, each answered briefly.
Part 1: Manual testing
What is the STLC, and how does it fit into the SDLC?
The software testing life cycle is the testing part of the software development life cycle. Its phases are requirement analysis, test planning, test case design, test environment setup, test execution and test closure. In agile teams these phases repeat every sprint, in small cycles, instead of running once.
Verification or validation?
Verification asks "are we building the product right?" It checks documents, designs and code against the specification, through reviews and walkthroughs, without running the software. Validation asks "are we building the right product?" It runs the software to check it meets the user's real needs. Testing a feature is validation. Reviewing its requirements is verification.
Test scenario, test case or test script?
A test scenario is a one-line goal: "Check login with valid credentials". A test case is the detailed version: preconditions, steps, test data and expected result. A test script is an automated test case, written in code or in a tool such as Tosca. Interviewers often ask you to turn a scenario into two or three test cases on the spot.
Severity or priority? Give an example of each combination.
Severity is how badly the defect breaks the system. Priority is how soon it must be fixed. High severity, low priority: the app crashes on a rarely used export to an old file format. Low severity, high priority: the company name is misspelled on the home page. The tester sets severity. The product owner or lead usually sets priority.
Smoke testing or sanity testing?
Smoke testing is a broad, shallow check on a new build: does it start, can you log in, do the main screens open? If it fails, the build goes back. Sanity testing is a narrow, deeper check after a fix: does this specific bug fix work, and did it break anything right next to it? Smoke decides if a build is testable. Sanity decides if a fix is acceptable.
Regression testing or retesting?
Retesting runs the failed test case again after a fix to confirm the defect is gone. Regression testing runs the wider set of tests to make sure the fix, or any change, didn't break something that used to work. Retesting is planned for specific defects. Regression is the usual candidate for automation, because it repeats every release.
Explain boundary value analysis and equivalence partitioning with an example.
Take an age field that accepts 18 to 60. Equivalence partitioning splits inputs into groups that should behave the same: below 18 (invalid), 18 to 60 (valid), above 60 (invalid), and one value from each is enough. Boundary value analysis tests the edges, where bugs cluster: 17, 18, 19, 59, 60 and 61. Used together, about eight test cases cover what would otherwise take dozens.
When do you use a decision table?
When the result depends on combinations of conditions. For example, a loan form where approval depends on income, credit score and existing loans. List every combination of conditions as columns, with the expected action for each. The table shows missing rules too. Developers often haven't decided what happens for one combination, and you've found a requirements gap before it became a bug.
Walk me through the defect life cycle.
New, then assigned to a developer, then open while they work on it, then fixed. The tester retests it: if it passes, it's closed, and if not, it's reopened. Along the way a defect can be rejected (not a bug), deferred (fix later) or marked a duplicate. Mention the tool you've used, such as Jira, and that your team's workflow may name the states differently.
What makes a good bug report?
A clear title that says what fails and where. Steps to reproduce, numbered. Expected result and actual result. Environment: build, browser, device, test data. Severity. Evidence: a screenshot, a screen recording or logs. A good test of a bug report: could a developer reproduce it without asking you a single question?
What is a requirements traceability matrix?
A table that maps each requirement to the test cases that cover it, and then to their results and any defects. It shows two things fast: requirements with no tests, and which requirements are affected when a test fails. In regulated work, such as banking or healthcare, it's often required for audits.
What goes into a test plan?
Scope, what's in and out, the test approach and types of testing, entry and exit criteria, environments and test data, schedule, roles, risks and how you'll handle them, deliverables and the defect management process. In agile teams the plan is often lighter, a page per release, but the same questions still need answers.
How would you test a login page?
Start with functional cases: valid login, wrong password, wrong username, empty fields, case sensitivity, and the "forgot password" flow. Then security: masked password, lockout after failed attempts, no credentials in the URL, session timeout, and SQL injection in the fields. Then usability and accessibility: tab order, error messages that don't reveal whether the username exists, and screen reader labels. Then compatibility across browsers and devices. Interviewers care about the order you think in more than the full list.
The developer says "it's not a bug". What do you do?
Go back to the source of truth: the requirement, the acceptance criteria or the design. If the behaviour contradicts them, show that with the evidence. If the requirement is unclear, it's not a fight between you and the developer. Take it to the product owner or business analyst to decide, and document the decision. Stay factual and keep the relationship good. You'll work with this developer next sprint too.
What is exploratory testing, and when is it useful?
Testing without pre-written cases, where you learn the application, design tests and run them at the same time. It's done in time-boxed sessions with a goal, called a charter, such as "explore the checkout with expired cards for 45 minutes". It finds bugs that scripted tests miss, especially in new features and edge cases, and works best alongside scripted regression, not instead of it.
How do you test an API without automation?
With a tool such as Postman. Check the status codes (200, 201, 400, 401, 404, 500), the response body against the contract, required and optional fields, invalid inputs, authentication, and how long responses take. Then check the effect in the database or UI. Knowing basic SQL to verify data is often what separates a strong manual tester.
What is your role as a tester in an agile team?
Testing starts before code exists. Join refinement to question unclear stories and help write acceptance criteria. Test within the sprint, not after it. Pair with developers on edge cases. Keep a lean regression suite and help automate the stable parts. Be clear on the definition of done, so "done" includes tested.
Part 2: Tricentis Tosca
What is Tosca, and how is it different from Selenium?
Tosca is a commercial, model-based test automation tool from Tricentis. You scan the application into reusable modules and build tests by combining them, mostly without code. Selenium is an open-source library: you write code to drive browsers. Tosca covers web, desktop, SAP, mobile and API testing in one tool, with test design and test data management built in. Selenium is free and very flexible, but needs programming and a framework you build yourself.
What are Modules, and how do you create them?
A Module is Tosca's model of one part of the application under test, such as a login screen, with its controls as ModuleAttributes. You create Modules by scanning the application with XScan, which captures each control and how to identify it. When the UI changes, you update the Module once, and every TestCase that uses it picks up the change. That reuse is the core of model-based testing.
What are ActionModes?
ActionModes tell a TestStep what to do with a ModuleAttribute. The main ones: Input enters a value, Verify checks a value, Buffer stores a value for later, WaitOn waits for a condition, Select acts on a control without changing its value, and Constraint narrows down a row in a table. Choosing the right ActionMode is a common interview check.
What is a buffer, and how do you use one?
A buffer stores a value during execution to reuse later, such as an order number created in step 3 and searched for in step 10. Write it with the Buffer ActionMode, and read it with the syntax {B[OrderNumber]}. Buffers make tests independent of hard-coded data. Clear or overwrite them deliberately, so one test doesn't depend on another's leftovers.
What is TestCase Design?
TestCase Design (TCD) separates test logic from test data. You define attributes and their instances, for example card type: Visa, Mastercard, expired card, and combine them using test design methods. Then a template TestCase is instantiated into one real TestCase per combination. Change the data in TCD, and the instances update. It's how Tosca covers many data variations without copying test cases.
What are Reusable TestStep Blocks and business parameters?
A Reusable TestStep Block is a group of steps, such as "log in", that you build once and use in many TestCases. Business parameters let each TestCase pass its own values into the block, such as a different username. Update the block once and every TestCase that uses it changes, much like a function in code.
What are recovery and cleanup scenarios?
A recovery scenario tells Tosca how to recover when a step fails unexpectedly, such as closing an unexpected pop-up and retrying, so one problem doesn't fail the whole run. A cleanup scenario returns the application to a known state, such as closing all browser windows and reopening the start page, when recovery isn't possible. Together they make long unattended runs reliable.
What are ExecutionLists?
An ExecutionList groups TestCases to run together, such as a regression suite or a smoke suite, and records the results of each run. You run them from Tosca Commander, from a CI pipeline, or distributed across agents for parallel runs. Results show pass and fail per TestCase and step, with logs and screenshots.
How do configuration parameters help?
Configuration parameters hold values that change by environment, such as the URL or browser. Set them on a folder or ExecutionList and read them with {CP[Url]}. Running the same tests on QA and then staging becomes a single change, not an edit in every TestCase.
How do you handle dynamic controls that change identification?
Re-scan with XScan and choose stable properties, such as a fixed ID, name or label, rather than ones generated at runtime. Use anchors to identify a control relative to a stable neighbour, or index-based identification as a last resort. Talk to the developers about adding test IDs. A good answer shows you prevent the problem, not just patch it.
How do you run Tosca tests in CI/CD?
Trigger ExecutionLists from the pipeline, for example Jenkins or Azure DevOps, through Tosca's CI integration and distributed execution agents. Results come back to the pipeline as pass or fail with reports. Keep a small, fast smoke suite on every build, and the full regression suite nightly. Mention which versions you've worked with. Tricentis lists current features in its Tosca 2025.1 LTS release notes.
How to prepare in the last 48 hours
Pick one real application, such as an e-commerce checkout, and practise writing 10 test cases for it out loud.
Prepare two defect stories in STAR form: a bug you found that others missed, and a disagreement you resolved. See STAR method examples.
For Tosca, be ready to explain one project end to end: what you automated, how you structured Modules and TCD, and how the suite ran in CI.
Brush up on basic SQL queries for checking data.
Practise these questions out loud with feedback in an AI mock interview. For live interviews, the InterviewPlan AI interview copilot can surface notes from your own prep on Zoom, Teams or Meet. Check your employer's rules on AI tools first.
