Simulated Interview: How to Run One on Your Own, Out Loud

A simulated interview is a full practice run of the real thing: questions for the role you are applying to, answered out loud, timed, and reviewed before you try again. Done alone, it takes about 30 minutes, and it works for one reason: it forces you to say answers you have only ever thought.

Most interview preparation is reading. You read the common questions, you read model answers, you nod. Then the interviewer asks “tell me about a time you disagreed with your manager” and you hear yourself saying “um, so, like” for twelve seconds while the story assembles itself. Reading did not fail you; it just trained a different skill. A simulated interview trains the one that is tested.

Photo by LinkedIn Sales Solutions on Unsplash

What makes a simulated interview different from “practising questions”

A simulated interview has four properties that casual practice usually skips:

1. Out loud. Answers spoken, not rehearsed in your head. Your head will happily skip the transition you cannot find words for; your mouth cannot.
2. One take per answer. No restarting the sentence. In the real room you cannot rewind, so in practice you do not either. Restart only after the answer is finished.
3. Measured. Something outside your memory of the answer tells you how it went: a recording, a timer, a person, or an app that counts words and pauses.
4. Repeated with one change. You fix a specific thing and run the same question again. Without the second run you have collected feedback, not used it.

Career services at universities run exactly this format with a counsellor playing the recruiter and a recording reviewed afterwards. You can rebuild most of it on your own.

The 30-minute solo loop

Block 30 minutes, close the door, put the phone on do-not-disturb, and work through this.

1. Pick five questions for the actual role (3 minutes). Two standard openers (“Tell me about yourself”, “Why this role?”), two behavioural questions (“Tell me about a time you…”), and one question specific to the job, such as a technical scenario or a customer situation. Write them down; do not write the answers.

2. Answer each one out loud, in one take (10 minutes). Sit as you would in the interview. Start the answer within three seconds of reading the question. Aim for about one to two minutes per answer. Record it, even if the recording is just your phone’s voice memo app.

3. Score each answer before you listen to it (2 minutes). Write down, from memory, how long you think it ran and how many times you said “um”. This sounds pointless. It is the most useful step, because the gap between what you think you did and what the recording shows is the thing you are training.

4. Review against three numbers (8 minutes). Listen back or read the transcript and check:

• Pace. Count the words in a 30-second stretch and double it. Roughly 110–150 words per minute reads as calm and clear. Under about 110 sounds like you are searching for the answer; over about 150 the detail blurs.
• Filler words. Count every “um”, “uh” and filler “like”. Two or three across five answers is normal. Ten is the habit you came here to break.
• Specifics. Does each behavioural answer contain a number (a percentage, a deadline, a team size, a result) and a verb that says what you did (built, cut, led, rewrote)? “We improved the process” is a story without a hero.

5. Write exactly three fixes and re-run (7 minutes). Not ten. Three, each tied to one answer: “Lead with the outcome in the conflict story”, “Replace um with a one-second pause”, “Add the 40% figure to the onboarding answer”. Then answer those same questions again and compare.

That is one simulated interview. Do it three or four times across the week before the real one, rather than once for two hours the night before.

What to listen for in the playback

The numbers above are the easy part. These four patterns are what usually separates a decent answer from a hire:

• The late outcome. You tell the whole story and mention the result in the final sentence, after the interviewer has stopped listening for it. Flip it: result first, then how.
• The borrowed “we”. Team language hides your part. Keep “we” for context and “I” for actions.
• The answer that ends twice. You finish, then add “so yeah, that’s basically it, I think it went well”. Stop at the result. Silence after an answer is fine.
• The memorised paragraph. If a recording sounds recited, the interviewer will hear it too. Prepare two or three core stories with their numbers, not scripts, and let the wording vary.

The STAR structure (situation, task, action, result) is a good default for behavioural questions, as long as the situation and task take one or two sentences and the action carries most of the answer.

Where a tool helps, and where it does not

The weakest step in the solo loop is step 4. Counting words per minute and filler words by hand is tedious enough that most people stop doing it by the second session, and then the practice goes back to being vague.

That counting is the gap we built Bussinly to close. You pick the role by industry, department and job title, answer questions out loud or in text, and see your pace and filler words live while you speak. At the end you get a score built from your pace, filler words, answer length and whether your answers contain numbers and action verbs, a verdict from Strong Hire to No Hire, and three things to fix from your most recent answer, which is the same loop as above with the counting done for you. Then you run it again and watch whether the number moves.

What a tool cannot do is play the interviewer who leans in after your answer and asks “why did you do it that way?” For that, recruit a friend for the last session before the interview: hand them your five questions and ask them to add one follow-up to each. A rule-based score tells you how your answer sounded; a person tells you whether it convinced them.

A one-week simulated interview plan

Day	Session	Focus
Day 1	30-minute loop, five questions	Baseline: record pace, filler count and missing numbers
Day 3	Same five questions	Apply the three fixes from day 1 only
Day 5	Five new questions, including one you dread	Speed of starting; one take each
Day 6	Full run with a friend	Follow-up questions, eye contact, ending answers cleanly
Day 7	15 minutes, two core stories	Warm-up only; no new material the day before

The plan is deliberately light. The goal is not to memorise answers; it is to make speaking under mild pressure feel familiar, so that in the real interview your attention goes to the question instead of to your own voice.

Simulated interview FAQ

What is a simulated interview? A simulated interview is a practice run of a real job interview: real questions for the role you want, answered out loud, under time pressure, with feedback afterwards. It differs from reading interview tips because you have to produce the answer in order, in your own words, without stopping to edit it.

Can I do a simulated interview on my own? Yes. You need a list of questions for the role, a way to hear or measure your answers (a voice recorder, or an app that scores pace and filler words), and the discipline to change one thing and run the same question again. A partner helps with follow-up questions, but most of the gain comes from repetition.

How many simulated interviews should I do before the real one? Three or four short sessions across the week before the interview beat one long session the night before. Each session should end with a short list of fixes, and the next session should start by re-running the questions those fixes came from.

What speaking pace should I aim for in an interview? Roughly 110 to 150 words per minute. Below about 110 most listeners hear hesitation; above about 150 the details start to blur. Pausing for a second instead of saying um or uh slows you down in the right way.

---

Bussinly is built by Teamz Lab, a small UK app studio. If you run a recruiting, training or coaching business and want a practice tool of your own, we take client builds through our Upwork agency profile, with milestone escrow.