This free interview question prep tool gives you the questions that actually get asked — grouped into eight sets, tagged by seniority — and a structured drafting space for each one, so you leave with written answers rather than a vague sense that you should probably think about this. Your drafts save in your own browser as you type, the word counter warns you when an answer has drifted past two minutes of speaking, and the download button exports everything as a plain text file you can read on your phone in the waiting room.
Arb Digital built this alongside our free resume tools because the two halves of a job application fail in different ways. A resume gets you into the room. What happens in the room is a separate skill, and it is the one almost nobody practises deliberately.
How to prepare for an interview without memorising scripts
Most interview advice tells you to rehearse answers. That is the wrong target. Rehearsed answers sound rehearsed, and the moment the question is phrased slightly differently, the script falls apart. What works is preparing material rather than answers: six or seven real situations from your own work that you know in detail, which you can reshape to fit whatever is actually asked.
The difference matters because interviewers rarely ask the question you prepared for — they ask a neighbouring one. You prepared "tell me about a time you led a project" and you get "tell me about a time you had to get something done without formal authority." A memorised script leaves you forcing the wrong answer into the right-sounding shape. Prepared material means you already have the situation in your head and simply emphasise a different part of it.
The STAR structure, and why interviewers use it
Behavioural questions all follow the same logic: past behaviour is the best available predictor of future behaviour. Interviewers are usually trained to score answers against four elements, which is why the STAR structure exists:
- Situation — the context, in one or two sentences. Where, when, what was at stake.
- Task — what you specifically were responsible for. Not the team, you.
- Action — what you actually did, step by step. This should be the longest part.
- Result — what changed, with a number wherever you have one.
The two elements people skip are Task and Result. Skipping Task makes it unclear whether you did the work or watched it happen. Skipping Result leaves the interviewer with a story and no outcome, which scores as neutral at best.
This is not an arbitrary convention. Structured interviewing — asking every candidate the same questions and scoring them against defined criteria — is standard practice in public-sector and large-employer hiring precisely because it is more consistent than unstructured conversation. The US Office of Personnel Management's guidance on structured interviews sets out how those scoring rubrics are built. When you know a rubric exists, the STAR fields stop feeling like a gimmick and start looking like the answer sheet.
Build a story bank, not a script
Before an interview, write out six to eight situations from the last few years, each with the STAR elements filled in. Choose them so they cover a spread: a success, a failure, a conflict, a deadline, something you led, something you learned. Almost any behavioural question can be answered from a bank like that, because interviewers are asking about the same handful of underlying qualities in different words.
A practical way to build the bank is to work backwards from the job posting rather than forwards from your memory. List the five or six qualities the posting keeps circling — ownership, cross-team coordination, working with ambiguity, customer focus — and then find one situation from your own history for each. That mapping exercise usually reveals a gap: a quality the role clearly cares about that you have no story for. Better to discover that at your kitchen table than thirty seconds into an answer. The bank is also what makes a second-round interview survivable, because if you have spent your three best stories on the hiring manager, the panel round needs different material.
The result is the part people get wrong
Most weak STAR answers are weak in the same place: the Result is either missing or unfalsifiable. "It went really well and the team was much happier" is not a result, it is an adjective. A result is something that changed and could in principle be checked — a release that shipped two weeks earlier, a backlog that fell from ninety tickets to twelve, a process that used to take three people and now takes one.
You do not need a spectacular number. You need a specific one. "We cut the onboarding call from ninety minutes to forty" beats "we dramatically improved onboarding," even though the second sounds bigger, because the first implies you were actually measuring. If you genuinely have no number — and plenty of real work has none — say what changed in concrete behavioural terms: what stopped breaking, what the team did differently afterwards. Never invent the figure, though: interviewers ask follow-ups, and a fabricated metric collapses immediately under "how did you measure that?"
Talk about failure properly
"Tell me about a time you failed" is not a trap, and the wrong answer is a disguised strength like "I care too much about quality". Interviewers are checking two things: whether you can recognise a real mistake, and whether you changed anything afterwards. Pick a genuine failure with real consequences, own your part in it plainly, and spend most of the answer on what you did differently next time.
There is a calibration problem here that catches people out. A failure too small — "I once sent a meeting invite for the wrong time zone" — reads as evasive. One too large or too recent, especially one still unresolved, leaves an open question about your judgement. The useful middle is a real mistake, a year or more back, that cost something measurable and whose fix you can describe in the past tense.
The salary question
If asked early for a number, it is reasonable to ask what range the role is budgeted for — many employers now publish it, and in a growing number of places they are required to. If you must give a number, give a researched range rather than a single figure, and base it on the role and market rather than on what you currently earn.
Do the arithmetic before the conversation, not during it. If the role is quoted hourly and you think annually, convert it first with our hourly to salary calculator; if it is quoted as a salary and you are comparing against contract work, the salary to hourly calculator runs it the other way. And remember that a gross figure is not what lands in your account — running the offer through our take-home pay calculator before you negotiate stops you from accepting a headline number that nets out lower than the job you already have.
When two offers are genuinely close on base pay, the decision usually turns on everything around it — bonus structure, employer retirement contributions, health premiums, time off, commute cost. The job offer comparison calculator puts both packages side by side so you are comparing totals rather than headlines, and the pay raise calculator tells you what a proposed increase actually is in percentage terms.
Questions they are not allowed to ask
Occasionally a question strays into territory an employer has no business exploring — marital status, plans to have children, age, religion, national origin, disability, and similar. In the United States these areas are governed by federal anti-discrimination law, and the EEOC's summary of prohibited employment practices is the plain-language reference for what is off limits. Other countries have their own equivalents.
Most of the time these questions come from an untrained interviewer making conversation rather than from deliberate screening, which is why the confrontational response is usually the wrong one. The practical move is to answer the legitimate concern hiding behind the question rather than the question itself. Asked whether you have young children, you can simply say you have no issues with the travel or hours the role requires. That closes the topic, gives them the reassurance they were fumbling for, and leaves you nothing you have to disclose.
Always have questions for them
"Do you have any questions for us?" is part of the assessment, not the wind-down. Having none reads as low interest. The strongest questions are specific to the conversation you have just had, and they tell you something you actually need to know: what success looks like in the first six months, why the role is open, how the team handles disagreement, what the interviewer finds hardest about working there.
The "Questions to ask them" set in the tool above is deliberately built around questions with diagnostic value rather than flattering ones. Asking why the role is open is not small talk — the answer distinguishes a growing team from a backfill from a revolving door, and the hesitation before the answer often tells you more than the answer does. Public career-service guidance, including CareerOneStop's interviewing resources from the US Department of Labor, makes the same point: the closing questions are an information-gathering opportunity you only get once.
Practical things that change outcomes
- Reread the job posting the morning of the interview. The vocabulary it uses is the vocabulary the interviewer will use.
- Have your resume in front of you. You will be asked about specific lines on it, and hesitating over your own history is costly.
- Prepare for the gaps. If your resume has a career break, a short tenure or a change of field, expect the question and have a calm two-sentence answer ready.
- For video interviews, test the setup on the actual platform, not a different one, and put the camera at eye level.
- Write down the interviewers' names when they introduce themselves.
- Time one answer out loud before you go. Almost everyone speaks longer than they think; a countdown timer running silently while you practise is the cheapest correction available.
Rehearsing out loud, not in your head
Reading a draft silently confirms that you wrote something sensible. It tells you nothing about whether you can say it. Almost every answer that feels tight on the page runs long out loud, and the parts that collapse are usually the transitions — the join between Situation and Task, and the join between Action and Result. Record one answer on your phone and play it back: you will hear the filler, the sentence that restarts three times, and the point about forty seconds in where you began explaining background nobody asked for. Fix that one habit and every answer improves, because it is usually present in all of them.
Related free tools from Arb Digital
Interview prep is the last step of an application, not the first. Before you get here, check the document that got you the call with our free ATS resume checker, which scores it against any job description, and build or rebuild it in the free resume builder — 28 templates with PDF, Word or image download, no account and no watermark. There are also resume examples for 150 job titles if you want to see how a strong version of your role reads.
On the money side, work out what an offer is really worth with the take-home pay calculator and the job offer comparison calculator, and if you are weighing employment against going independent, the freelance hourly rate calculator works out the rate you would need to charge to match a salary. For roles that test speed at a keyboard as part of hiring, our typing speed test is free and unlimited. Browse the full free online tools hub for everything else.
Frequently asked questions
They save in your own browser's local storage so you can come back to them. Nothing is uploaded, nothing is associated with you, and clearing your browser data removes them. That is also why there is no login.
Ninety seconds to two minutes for a behavioural question. Much shorter and you have not given enough detail to score; much longer and the interviewer stops tracking the thread. Reading a draft aloud is the fastest way to check.
Try not to. If you keep returning to one project, the interviewer learns about one project rather than about you. That is exactly why a bank of six to eight situations is worth building.
STAR stands for Situation, Task, Action, Result. It is the four-part structure interviewers are commonly trained to score behavioural answers against: the context, what you specifically were responsible for, what you did, and what changed as a result. The Action section should be the longest part of your answer.
Prepare six to eight situations rather than a fixed number of questions. Because most behavioural questions probe the same underlying qualities in different words, a well-chosen bank of situations covers far more ground than the same effort spent scripting individual answers.
Say so plainly and offer the closest thing you do have, framed as such. Inventing an example is far riskier, because follow-up questions expose it quickly. An honest "I have not faced exactly that, but here is the nearest situation" scores better than a story that falls apart under one question.
Yes. The "Download my answers" button exports every question you have drafted an answer to as a plain text file, with the STAR fields labelled. Nothing is sent to a server — the file is generated in your browser and saved directly to your device.
Completely free, with no sign-up, no account and no limits. Work through as many question sets as you like, as often as you like. Everything runs in your browser and nothing you write is stored or transmitted anywhere.
Guidance on this page is general career information for planning purposes only, not legal or professional employment advice. Employment rules and interview practices vary by country, state and employer.
