How Long Should an Interview Answer Be?

There's no single right number of seconds. The honest answer is a range, and it depends on the question. Indeed's career-advice guidance puts most interview answers somewhere between 30 seconds and four minutes: short and direct for simple factual questions, longer for behavioral questions that need a real story (Indeed). For a structured behavioral answer using the STAR method (Situation, Task, Action, Result), the commonly cited target is tighter: roughly a minute and a half to two minutes (BetterUp).

So the range is real, and it's wide enough that memorizing a number won't help you much. What actually determines whether an answer feels too long isn't the clock. It's where your point lands inside it.

The problem usually isn't length. It's order.

Picture two versions of the same answer to "tell me about a time you improved a process." One spends forty-five seconds setting the scene (the team, the tool, the backstory) before finally saying what changed. Another says what changed in the first sentence, then spends the next forty-five seconds on the same details, now offered as support for a point the interviewer already has.

Both answers can be the same length. Only one of them feels long. An interviewer who has heard the point already has somewhere to put the details that follow: they're evidence. An interviewer who's still waiting for the point is doing something closer to guessing where the story is headed, and guessing reads as rambling even when the content is fine.

This is the actual lever: say the point first, then give the supporting reason. Not "here's everything that happened, and therefore here's my point," but the reverse. State the conclusion, then back it up. It costs you nothing in content. It changes everything about how the length feels.

A structure that scales to any length

"Point first, then one reason" isn't just an opening trick. It holds up whether your answer needs to be twenty seconds or two minutes:

• The point. One sentence. What do you want them to remember if they remember nothing else?
• One reason or example. The concrete thing that makes the point credible: a number, a specific moment, a result.
• Stop. If they want more, they'll ask a follow-up question. That's a good outcome, not a sign your first answer was incomplete.

Notice what's missing: a mandate to cover every detail of the situation. The Situation part of STAR is where most answers get long, because it feels safe to over-explain the setup before getting to the part that's actually about you. Keep it to one or two sentences and trust the listener to ask if they need more context.

A 60-second practice method you can do today

You don't need an app, a partner, or a mirror for this. You need your phone's voice recorder and about five minutes.

1. Pick one real question. Not a random one. Pick one you're actually likely to get: "Tell me about yourself" or "tell me about a time you had to convince someone" both work.
2. Set a timer for 60 seconds and hit record. Answer out loud, at a normal pace, like you're actually in the room. Don't stop early if you finish sooner. That's useful information too.
3. Before you listen back, write down what you think your point was, in one sentence, from memory. This is the part people skip, and it's the most useful step: you're checking whether your intended point matches what you actually said.
4. Now play the recording and note the timestamp where you first said something that sounds like that point. Not implied it. Said it, in words a stranger could repeat back.
5. Compare the timestamp to your total answer length. If your point landed in the first 15 seconds of a 60-second answer, you're in good shape regardless of the total length. If it landed at second 40, you've found the actual problem, and it isn't that the answer was too long. It's that the useful part was too short relative to everything before it.

Do this with three or four different questions in one sitting and a pattern usually shows up fast: most people have one or two openings they default to ("So, I guess I'd say...", "That's a good question, let me think...") that quietly eat the first ten or fifteen seconds before anything answerable begins. Once you can hear that opening, it's easy to cut.

What this looks like measured, not guessed

The exercise above works with a stopwatch and your own judgment about "does this sound like the point." That's genuinely enough to improve on your own. It's also exactly the thing we built the speaking line to show automatically: your answer drawn as one line, with the stretch where your point actually landed marked heavier than the rest, so you don't have to guess at the timestamp by ear. Same idea as the exercise above, just measured instead of estimated, and compared against your own baseline instead of a fixed target.

Practice with a stopwatch first. If you want it measured for you, that's exactly what we built.