How did recent UPSC toppers use mock tests in their preparation?

Verified topper accounts consistently show three patterns: sustained mock volume (35–45 full tests) in the final preparation year, rigorous post-test analysis rather than score-chasing, and progressive difficulty — starting with topic-wise tests then scaling to full simulations. Kanishak Kataria (AIR 1, 2018), Shubham Kumar (AIR 1, 2020), Shakti Dubey (AIR 1, 2024), and Anuj Agnihotri (AIR 1, 2025) all documented systematic test-and-analyse strategies.

Verified Topper Mock Test Strategies

Kanishak Kataria — AIR 1, CSE 2018

Background: IIT Bombay, Mathematics and Computing graduate. Secured AIR 1 in his first attempt.

Mock approach: Kataria brought an engineering-discipline framework to Prelims preparation. He followed a strict test → analyse → revise loop — taking a mock, meticulously reviewing every wrong answer against source material, scheduling targeted revision of identified gaps, and only then taking the next mock. He did not rush through tests or treat volume as the goal. He combined multiple question sources rather than relying on any single coaching series, which exposed him to broader question diversity.

Key insight: Kataria's approach was data-driven — he tracked which topics produced wrong answers across multiple tests and treated that data as a directed revision schedule, not as a score report card.

Shubham Kumar — AIR 1, CSE 2020

Background: IIT Bombay graduate. Cleared CSE 2020 in his third attempt.

Mock approach: Shubham Kumar explicitly and repeatedly credited his increase in mock volume as one of the key differences between his unsuccessful 2nd attempt and his successful 3rd attempt. He went from approximately 20–25 mocks in his 2nd attempt to 40–45 mocks in his 3rd, with detailed post-test analysis after each. He also stated, in multiple published interviews, that:

Mock scores should never be used as predictors of actual performance — only as diagnostic tools
The analysis phase (finding out why each wrong answer was wrong) was more valuable than the test-taking itself
He took mocks at the actual exam time (9:30 AM) for conditioning

What changed between attempts: Not just the volume — the analysis rigour increased proportionally. 40 mocks with shallow review produced less improvement than 40 mocks with deep analysis.

Aditya Srivastava — AIR 1, CSE 2023

Background: Electrical Engineering; cleared in first attempt.

Mock approach: Srivastava emphasised current affairs integration in mock test selection — he looked for a series that linked static syllabus questions to recent events, aligned with UPSC's post-2019 pattern shift toward application over pure factual recall. A test that asks whether you can connect the RTI Act to a specific transparency controversy is more exam-representative than one that only asks the year the Act was passed.

Shakti Dubey — AIR 1, CSE 2024

Background: Post-graduate in Biochemistry, BHU; PSIR optional. Cleared in her fifth attempt (failed Prelims in attempts 1–3; cleared Mains in attempt 4 but did not make the final list; AIR 1 in attempt 5).

Mock approach: Joined test series for both Prelims and Mains. Practised PYQs systematically. Used limited, curated resources — avoiding information overload. Most distinctively: treated each unsuccessful attempt as the most realistic mock available and calibrated her next preparation cycle based on that real-world diagnostic data, not on mock series averages alone.

Anuj Agnihotri — AIR 1, CSE 2025

Background: MBBS from AIIMS Jodhpur; Medical Science optional. Cleared in his third attempt through 13 hours of daily self-study. Did not join a full coaching institute — took only NEXT IAS's CA-VA current affairs course and Legacy IAS's Interview Guidance Programme (IGP).

Mock approach: Integrated PYQ practice, mock tests, and current affairs throughout preparation. His high interview score (204/275) was explicitly linked to structured mock interview practice at Legacy IAS's IGP — demonstrating that mock discipline applies to the Personality Test stage, not just Prelims. Verified Mains marksheet: Essay 108, GS-I 111, GS-II 127, GS-III 103, GS-IV 126, Medical Science I 142, Medical Science II 150.

Common Patterns Across All Four Toppers

Pattern	Detail
Volume	35–45 full-length Prelims mocks in the final preparation year (Shubham Kumar's explicitly verified count)
Analysis rigour	2–3 hours of post-test analysis for each 2-hour test
PYQ priority	All four documented systematic PYQ practice alongside coaching series mocks
Timing conditioning	Tests taken in the actual exam time slot (9:30 AM) for at least some sessions
Strategy adaptation	Changes were data-driven (from mock analysis patterns) not emotion-driven (from single bad scores)
Selective coaching	None used every service from a single institute — all were selective and targeted

The Universal Warning From All Four

Every documented account from these toppers — regardless of background, optional subject, or number of attempts — includes the same warning:

Do not use a single mock score as a measure of preparation quality or as a predictor of exam outcome. Score trends across 15–20 tests — and the quality of analysis after each test — are the meaningful signals.

Shakti Dubey's five-attempt journey illustrates this most powerfully: failed Prelims in attempts 1, 2, and 3; cleared Mains in attempt 4 but did not make the final list; AIR 1 in attempt 5. Each attempt functioned as the most realistic diagnostic mock available — and she calibrated each subsequent preparation cycle using that data. The score of any single attempt or mock is not the end of the story.

Three things all four toppers had in common: (1) analysis time per mock matched or exceeded test time; (2) score trends — not single scores — guided strategy changes; (3) PYQs were practised extensively alongside any coaching series mocks.