Grade your interview answer in seconds
Paste the question and your answer. You get a 0–100 score, the strengths and gaps, and a rewrite in the structure the question calls for. No sign-up needed.
How to use the grader
- Paste the exact question. "Tell me about a time you disagreed with your manager" is graded differently from "Why do you want this job?", because the question type decides the criteria.
- Write the answer the way you would say it. Aim for 150 to 300 words, about one and a half to two minutes spoken. Very short answers score low on specificity; very long ones lose structure.
- Read the lowest sub-score first. That is usually the fastest improvement. A missing Result or a vague Action costs more points than wording.
- Compare your answer with the rewrite. The rewrite keeps your story. Notice where it adds numbers, cuts context, or turns "we" into "I".
- Say it out loud. An answer that reads well can still ramble when spoken. Practicing aloud is where the score starts to stick.
What the score looks at
Every answer is first classified into one of six question types: behavioral, technical, situational, introduction, motivation, or strengths and weaknesses. Each type has five criteria. For a behavioral question those are STAR structure, specificity, measurable impact, relevance and authenticity. For "Tell me about yourself" they are structure, relevance, conciseness, engagement and positioning. The full rubric, and its limits, is on ourscoring methodology page.
Scores map to four bands: 80 and above is excellent, 60 to 79 good, 40 to 59 fair, and below 40 needs work. Most first drafts land in the 50s and 60s. Getting to the 80s usually takes a concrete result and a tighter Action.
An example
Question: "Tell me about a time you had to meet a tight deadline."
“We had a big project due and everyone worked really hard. I stayed late a few nights and we got it done in the end. It was stressful but we made it.”
Why it falls flat: No specifics, no personal action, no result beyond 'we made it'. It would score around 30.
“Our client moved a product launch forward by a week, so the onboarding emails I owned had five days instead of twelve. I cut the sequence from seven emails to the four that drove sign-ups last quarter, got legal to review all four in one session instead of four, and scheduled QA for day four. We launched on time, and the shorter sequence converted 12% better than the old one.”
Why it works: Clear situation and task, three concrete actions owned by the speaker, and a measurable result. It would score in the 80s.
The grader would rewrite the weak answer into STAR along these lines:
Situation
A client moved our product launch forward by a week.
Task
I owned the onboarding emails, which now had five days instead of twelve.
Action
I cut the sequence to the four emails that performed best, batched the legal review into one session and booked QA a day early.
Result
We launched on time and the shorter sequence converted 12% better.
These examples are written for this page; they are not real candidates' answers.
Common reasons answers score low
- Too much setup. If the Situation takes half the answer, the Action is squeezed. Keep context to two sentences.
- "We" instead of "I". Interviewers are hiring you. Say what you personally did.
- No result. End with an outcome: a number, a decision, a change that stuck, or what you learned.
- Answering a different question. A great story about teamwork does not answer "Tell me about a mistake".
From a written answer to a spoken one
The grader checks the content of an answer. Interviews also test delivery: pace, filler words, and how you handle a follow-up question you did not expect. In Offermic you answer out loud to an AI interviewer who asks those follow-ups, and every answer is scored the same way. Start with the grader, then rehearse the answers that matter most out loud.
Questions people also ask
Is the answer grader free?
Yes. You can grade 3 answers a day without an account. A free account raises that to 10 scored answers a day and adds a weekly live mock interview.
How is my answer scored?
The grader works out the question type, then rates five criteria that fit it, such as STAR structure, specificity and measurable impact for behavioral questions. The ratings add up to a score from 0 to 100.
Do you keep my answer?
Your text is sent to our AI provider to be graded. Anonymous results are not added to a practice history. See the privacy policy for details.
Can it grade answers in other languages?
Yes. Write in any of the 13 languages the app supports and the feedback comes back in the language of this page.
How accurate is the score?
It is an AI judgement, useful for spotting weak structure, vague claims and missing results. It is not a prediction of whether you will get the job, and the same answer can move a few points between runs.