Types of Interview Questions and What Each One Tests

Interviewers rarely announce which kind of question they are asking, but almost everything you will hear belongs to one of six families. Each family exists because it measures something the others cannot.

• Behavioral asks for a real past event: tell me about a time you disagreed with your manager. It tests evidence, not intent.
• Situational asks for a hypothetical: what would you do if a release broke on a Friday night. It tests judgment when you have no track record to point at.
• Competency targets one named skill against a rubric: walk me through how you prioritize a backlog. It is often scored on a scale the interviewer is holding.
• Technical asks you to produce something correct: write the function, query the table, design the schema. Process counts, but a wrong answer is still wrong.
• Case and estimation hand you an ambiguous business problem and watch you structure it. Consulting uses these heavily; product and data roles borrow them.
• Culture-fit and motivation covers why this company, why now, what you want next. It sounds like conversation and is scored like everything else.

Question banks organized by role and round sit on the question banks hub if you want the actual sets rather than the taxonomy.

Losing ten seconds deciding whether a question wants a story or a framework reads as hesitation, and it happens to prepared candidates constantly. The tells that identify the type are reliable, and nearly all of them sit in the first four words. Here is what to listen for.

• Past tense plus a time marker (tell me about a time, describe a situation where) means behavioral. Answer with one specific event.
• Conditional mood (what would you do if, imagine that) means situational. Answer with a decision process, then the decision.
• How do you in the present tense means competency. They want your standing method, not one anecdote.
• An imperative (write, design, walk me through this code) means technical. Clarify constraints before you start producing.
• A number or a market (how many, should this company launch) means case or estimation. State your structure out loud before any arithmetic.

One caution: interviewers mix types inside a single question. Tell me about a time you handled a production outage, and what would you change now is behavioral followed by situational. Answer both halves, in that order, or the follow-up will drag you back.

Using the wrong structure is the most common avoidable mistake in interviews, because a good answer in the wrong shape still scores poorly. Behavioral and situational questions reward STAR, which is covered in detail on our STAR method guide. Competency questions reward a stated method followed by one short proof point. Technical questions reward thinking aloud, naming your assumptions, and checking edge cases. Case questions reward a visible structure before any number.

A concrete version: a backend engineer interviewing for a senior role at a payments company is asked how do you approach on-call. That is competency, present tense, and the trap is to answer it as a war story. The strong answer states the method first (runbooks, alert thresholds, blameless review), then spends two sentences on one incident as evidence. The war-story-only answer gets the follow-up and is that how you always work?, which is the interviewer telling you they did not get what they asked for.

Culture-fit questions have their own failure mode. They are the ones candidates answer honestly and casually, because the tone drops. The scoring does not drop with it.

Running the same question through two different structures is the fastest way to feel the difference; the mock interview tool will ask follow-ups the way a real interviewer does.

Memorised answers break the moment a question changes type, which is exactly what follow-ups do. Practicing by type is more durable: take four or five of your own strongest experiences, then rehearse each one as a behavioral story, as a competency method, and as the raw evidence behind a situational answer. The material stays the same; only the shape changes.

SubcueAI can help on both sides of that. The mock interview practice mode asks questions and follows up on your answers, and during a real interview the desktop app and the browser extension side panel can surface a suggested structure on a local overlay while you talk. What neither can do is supply the specifics. The tool can tell you that a question is behavioral and that STAR fits; the outage you handled and the number you moved have to be yours.

Honest limits apply to the live side. If the interview is proctored, recorded, screen-shared, or running on a company-managed device, assume anything on your screen is in scope. More question sets by role and round are on the question banks hub, and format-specific guidance sits under interview types.