After a rejection: ask for feedback and keep a review log
In this article (6 sections)
Related programme: the Data Science course (6 months; ₹1,30,000 (inclusive of GST) · EMI from ₹5,000/month).
After a job rejection, send a short, specific feedback request within a day or two, accept that many companies will not reply, and record every rejection in a review log with the round reached and your own notes. The feedback you receive will be patchy; the log is what turns ten rejections into one clear pattern you can fix.
The feedback request
Generic requests ("any feedback would be appreciated") are easy to ignore. Ask one narrow question that a busy interviewer can answer in a sentence.
Subject: Data Scientist interview - thank you - [Your Name]
Hi Aditi,
Thank you for letting me know, and for arranging the interviews. I understand
the decision.
If it is possible to share, I would find it useful to know whether the main gap
was in the modelling case discussion or in the SQL round. One line would help me
prepare better.
Regards,
[Name]Why this works:
- It accepts the decision up front, so there is no risk of an argument.
- It names the two rounds, so the recruiter can forward it to the panel with a simple question.
- "One line" sets a small expectation.
Send it to the recruiter who informed you, not to interviewers directly, unless you had direct contact with them during the process.
What to expect back
Many companies have policies against detailed feedback, and many recruiters simply do not have it. Common replies, and what they mean:
| Reply | What to take from it |
|---|---|
| No reply | Normal; log it and move on |
| "We went with a candidate with more experience" | Could be true; check whether the JD asked for more years than you have |
| "Strong candidate, close decision" | Worth applying again later; note the date |
| A specific comment on a round | Gold; log it word for word |
The review log
Add a second sheet to your job application tracker, with one row per rejection.
| Column | Example | Why |
|---|---|---|
| Date | 12 Oct 2026 | Order |
| Company type | Startup | Patterns by type |
| Role | Data Scientist | |
| Round reached | Technical 2 (case) | Where you drop out |
| Questions asked | Churn model; how to pick a threshold | Raw material for practice |
| My self-rating | 2 of 5 | Honest view straight after |
| What went wrong, in my view | Did not discuss precision/recall trade-off | Your own diagnosis |
| Feedback received | "Case discussion lacked business framing" | External view |
| Action | Practise 3 case answers with a business metric first | What changes |
Fill in the self-rating and "what went wrong" on the day, before any feedback arrives. It stops the recruiter's reply from overwriting your own memory of the interview.
Reading the log after ten rows
Sort by round reached. A fictional example after ten rejections:
Round reached Count
Resume screen 4
Online test 0
Technical 1 1
Technical 2 (case) 4
Manager round 1This candidate's problems are in two places: the resume is not getting through for some roles, and case rounds are failing. The test and first technical round are fine. So the plan is not "study everything"; it is:
- 1For the four screen rejections, compare the resume against those JDs using the method in testing your resume against job description keywords.
- 2For the four case rejections, collect the questions from the log and practise answering each with the business metric first, then the method.
Protecting your morale
Rejections in a job search are normal. A few habits help:
- Limit how often you check email for results. Once in the morning and once in the evening is enough.
- Count activity, not outcomes, in your weekly review: applications sent, practice sessions done, projects improved.
- Talk to someone who has hired, not only to other candidates. One conversation with a hiring manager about what they look for can reframe several rejections.
When to reapply
If the feedback was "close decision", a polite note to the recruiter three to six months later, with one concrete thing you have improved, is reasonable:
Hi Aditi, we spoke in October about the Data Scientist role. Since then I have
completed a pricing elasticity project and published it on GitHub. If a similar
role opens, I would be glad to be considered.If the log keeps pointing to the same technical gap in modelling rounds, a structured course can close it faster than practice alone. The Data Science programme runs for six months with live online sessions and recordings; the free skill check is a quicker first step.
Pankit Kumar has 10 years in Data Science & AI, building and shipping production systems in regulated pharma and clinical environments. He is a freelance trainer at Boston Institute of Analytics, AnalytixLabs and Scaler, and has taught this material to thousands of working professionals.
This article is part of our Data Science programme — 6 months. From data foundations to machine learning, deep learning and deployment.
Explore Data Science