For the first two months of my job search, every interview felt like starting from scratch. I would do a first round, it would go okay, I would not get feedback, and I would move on. No record. No reflection. No iteration. Just vague memories of what I thought went well and a general feeling that I "needed to get better."
Getting better at something you are not systematically measuring is very slow. It took me three months to realise I was making the same mistake in every phone screen — and I only realised it because I started logging every round in Gaply's Interview Tracker and the pattern became impossible to ignore.
What You Log and Why It Matters
For every interview round, the tracker captures:
- Company and role
- Round number and type (phone screen, technical, behavioral, final)
- Interview date and interviewer name
- Outcome: Cleared, Not Selected, or Awaiting Result
- Self-rating: 1–5 stars for how you felt you performed
- What went well — specific points from the conversation
- What to improve — specific things you would do differently
- Additional notes — anything else worth remembering
This takes about five minutes to complete immediately after an interview, while everything is fresh. Those five minutes are some of the highest-leverage time in an entire job search.
The Pattern I Would Never Have Seen Without the Data
After logging 12 rounds across eight companies, I looked at the summary statistics. My overall pass rate was 38%. My self-ratings averaged 3.2 out of 5. And then I looked at the "What to improve" section across all 12 entries.
The same phrase appeared seven times in slightly different forms: "Gave too much context before the point," "Took too long to get to the answer," "Spent too much time on background, not enough on the outcome."
I was burying the headline. In every interview, I was starting with extensive context and ending with the result. Interviewers, by the time I got to the actual answer, had often checked out or lost the thread. The fix was simple: lead with the conclusion, then explain how I got there. BLUF — Bottom Line Up Front.
I changed this in my next four interviews. My pass rate went from 38% to 71% over the following six weeks. Same experience. Same skills. Completely different structure in how I delivered my answers.
The Self-Rating Discrepancy
One of the more interesting insights was the gap between my self-ratings and my actual outcomes. I gave myself a 4/5 for two interviews I failed and a 2/5 for one interview I passed. My subjective sense of how it went was a poor predictor of the actual outcome.
Tracking this discrepancy over time helped me calibrate. When I felt confident after an interview (high self-rating) it often meant I had been comfortable, not effective. The most productive interviews often felt harder — more pushback, more follow-up questions — which I was misreading as going badly.
That recalibration changed how I interpreted real-time signals during interviews. Instead of getting rattled by tough follow-up questions, I learned to see them as interest, not attack.
Using the Data to Prepare for Future Rounds
Before any interview, I now review every previous entry for that company and for similar company types. I can see exactly what went well in my previous phone screen at a similar-sized startup, what I said in my technical round that the interviewer pushed back on, and what questions caught me off-guard.
This turns every new interview into a continuation of an ongoing investigation rather than a cold start. The data from round one directly improves my preparation for round two.