Getting Feedback and Improving Your Answers
Practice without feedback tends to reinforce whatever habits you already have, good and bad. This lesson focuses on turning your mock interview sessions into genuine improvement by using AI to give you specific, structured, and honest feedback, and on how to tell the difference between feedback that is actually useful and feedback that is just vague encouragement or vague criticism dressed up as analysis.
Why Generic Feedback Is Not Enough
If you ask an AI tool “how did I do,” you will often get a pleasant, generic response along the lines of “you did well, your answers were clear and confident.” This feels good but teaches you nothing, and worse, it can create false confidence. The fix is to ask for feedback against specific, named criteria rather than an open-ended judgment. A far more useful prompt is: “Review my last five answers against these specific criteria: did I use a clear structure like STAR where relevant, did I lead with the point instead of burying it, did I include a specific measurable result, was my answer an appropriate length for the question, and did I avoid vague or generic language. Score each answer against these criteria and tell me my single biggest area to improve.” Naming the criteria in advance forces the AI to give you something you can act on.
Reviewing Transcripts, Not Just Impressions
If you conducted your mock interview in text, you already have a full transcript to review, which is one of the most valuable things AI-assisted practice gives you that traditional practice with a friend rarely provides: an exact, unedited record of what you actually said. If you practiced out loud, transcribe your recording yourself or use a transcription feature if your tool offers one, and paste the transcript in for analysis afterward. Ask: “Here is the transcript of my mock interview. Identify the three answers that were weakest, explain specifically why, and rewrite one of them as an example of how it could be stronger.” Working from your actual words, including your filler phrases and tangents, produces far more useful feedback than working from your polished intended version.
Common Patterns Worth Asking About Directly
Certain weaknesses show up repeatedly across candidates and are worth checking for explicitly rather than waiting for the AI to volunteer them. Ask directly: “Did I use filler phrases like ‘um,’ ‘so basically,’ or ‘I guess’ excessively in this transcript? Did any of my answers run too long, over two minutes, without a clear reason? Did I ever answer a different question than the one that was actually asked?” This last pattern, subtly drifting away from the actual question asked, is surprisingly common and hard to notice in yourself without a structured review, because in the moment your answer feels responsive even when it has wandered.
| Criterion | Question to ask | What good looks like |
|---|---|---|
| Structure | Did I lead with the point before the detail? | Clear headline answer, then supporting detail |
| Specificity | Did I include real numbers or concrete detail? | Named outcomes, not vague claims of success |
| Length | Was the answer an appropriate length? | Most answers land between 60 and 120 seconds |
| Relevance | Did I actually answer the question asked? | No drifting into a different, easier topic |
| Delivery | Did filler language reduce clarity? | Minimal “um,” “so,” “I guess,” confident phrasing |
Iterating Instead of Starting Over
When feedback identifies a weak answer, resist the temptation to discard it and write something completely new from scratch. Ask your AI tool to revise the existing answer with the specific feedback applied: “Rewrite my answer to the conflict-resolution question, keeping the same underlying story, but leading with the outcome first and cutting the background detail by half.” Iterating on real material you have already rehearsed is faster and produces something you can actually deliver naturally, compared to memorizing an entirely new answer under time pressure close to the interview date.
Tracking Improvement Across Sessions
Keep a simple running log in your Interview Prep folder noting, after each mock session, your single biggest weakness identified and what you changed in response. This turns feedback into a visible trend rather than a one-off correction, and it is genuinely motivating to look back after four or five sessions and see filler language drop away or answers become consistently tighter. Ask your AI tool periodically to review this log itself: “Here is my feedback log from the last four mock interviews. What pattern do you see across sessions, and what should I focus on in my next practice round?” This kind of longitudinal review is something that is genuinely hard to get from a single well-meaning friend running one mock interview with you, and it is one of the clearest advantages of AI-assisted practice over traditional methods.
Knowing When You Are Ready
You do not need every answer to be flawless before you feel prepared. A more realistic bar is this: you can deliver your opening answer and your five to seven core STAR stories smoothly without notes, you have practiced at least one full mock interview under each of the interviewer styles covered in the previous lesson, and your feedback log shows the same one or two weaknesses have genuinely improved rather than persisted. Once you reach that point, shift your remaining preparation time toward the next lesson: preparing thoughtful questions to ask the interviewer, which is an equally important, often under-prepared part of the conversation.
