Updated April 2026 · 11 min read · India's most widely used campus assessment platform
| Company | Mercer Mettl (acquired by Mercer in 2018) |
|---|---|
| Headquarters | Gurgaon, Haryana, India |
| Market | Dominant in India campus recruitment · Expanding in Southeast Asia and Middle East |
| Products | Cognitive Ability Tests · Technical Assessments · Coding Tests · Personality Questionnaires · Proctored Exams |
| Major clients | Infosys, Wipro, HCL, Mahindra, HDFC Bank, Bajaj, Bosch, Honeywell, Cognizant |
Mercer Mettl is India's most widely deployed online assessment platform for campus and lateral hiring. If you're applying to a large Indian corporation — in IT, banking, manufacturing, or FMCG — there is a strong probability your online test is delivered through the Mettl platform, even if the test is branded to the employer. Unlike SHL or Kenexa (which design and own their test content), Mettl is primarily a delivery platform — employers can build custom tests or use Mettl's own content library on the same platform.
Standard aptitude tests covering quantitative ability (percentages, ratios, data interpretation, time-work, profit-loss), logical reasoning (series, puzzles, blood relations, syllogisms), and verbal ability (comprehension, sentence correction, vocabulary). Duration varies by employer configuration — typically 60–90 minutes for a combined aptitude test. No negative marking on most employer configurations, but some add negative marking for final-round assessments. Difficulty calibration varies — Mettl-delivered tests for Bosch or Honeywell are noticeably harder than those for smaller employers.
Role-specific technical MCQs in areas including: Java, Python, C++, SQL, data structures and algorithms, system design, cloud computing (AWS, Azure), cybersecurity basics, and embedded systems. Used heavily by IT companies (Infosys, Wipro, Cognizant, HCL) to screen for technical knowledge beyond what coding tests reveal. Output prediction questions (given code, what is printed?) and debugging questions (identify the error in this code snippet) are the most common formats.
Browser-based coding editor supporting C, C++, Java, Python, and sometimes JavaScript and Scala. Candidates solve 1–3 problems (Easy to Hard) in 30–60 minutes. Test cases are partially visible. Partial marking is awarded for passing a subset of test cases. Time complexity matters on harder problems — brute force solutions may time out on large hidden test cases.
Mettl offers Big Five personality assessments, cultural fit questionnaires, and sales aptitude tests. Format varies by employer but typically involves rating statements on a 5-point Likert scale. Some employers use forced-choice format. Scoring is against employer-defined profiles — there are no universally "correct" answers.
Mettl's proctoring system is among the most stringent of any assessment platform. It monitors via webcam, flags tab switches, detects multiple faces, and logs all mouse movement. Some employers enable "AI proctoring" — the system automatically flags suspicious behaviour for human review. Key rules: do not switch browser tabs under any circumstances. Do not leave the camera frame. Do not allow others to appear in your background. Use a stable internet connection — drops mid-test may cause automatic submission. Check the proctoring requirements in your invitation email before starting.
Because Mettl is a delivery platform rather than one fixed test, the specific combination you encounter depends entirely on the employer and role, and can span anywhere from a single 30-minute cognitive test to a multi-stage battery combining cognitive, technical, coding, and personality components across several hours. Your invitation email is the most reliable source — it typically lists the sections and their durations, and sometimes names a specific configured test (Mettl allows employers to name their custom assessment however they like, so you might see a branded name like "TechCorp Graduate Assessment 2026" that gives no obvious clue it's Mettl-powered underneath). If the branding is unclear, checking whether the test URL contains "mettl.com" or "mercer.com" is a reliable way to confirm you're on the Mettl platform even when the employer's own branding obscures it.
Unlike the coding round, which tests whether you can produce working code, the technical MCQ section tests conceptual knowledge directly — output-prediction questions ask you to trace through a code snippet mentally and determine what it prints, while debugging questions ask you to spot the specific error in a snippet without running it. Both formats reward genuine familiarity with a language's syntax quirks and common pitfalls (operator precedence, off-by-one errors in loops, mutable-default-argument traps in Python) more than they reward deep algorithmic knowledge. Since there's no code editor to test your answer in, building the specific skill of tracing code execution mentally — rather than relying on running code to check your intuition, which is how most developers normally work — is worth deliberately practising in the days before your test, since it's a different mode of thinking than day-to-day coding.
Because Mettl's coding assessments award partial marks for passing a subset of hidden test cases rather than requiring a fully correct solution to score anything, a submission-strategy decision worth making deliberately: on a problem you're not confident you can fully solve within the time limit, submitting a partially-correct but genuinely working solution — one that handles the common cases correctly even if it misses edge cases — usually scores meaningfully better than leaving the problem unattempted, or than submitting a more "complete-looking" solution that actually fails to compile or crashes on basic input. Time-complexity awareness matters specifically because hidden test cases on harder problems are often deliberately large enough to time out a brute-force approach that would pass all the small, visible test cases — a solution that looks correct against what you can see can still score zero on the hidden cases if its underlying approach doesn't scale, which is why testing your own solution mentally against a much larger input size than the visible examples show is worth doing before submitting, even without being able to actually run that larger test yourself.
Mercer Mettl tests are designed to be fast and broad — 12 to 50 questions in 10 to 20 minutes, mixing numerical, verbal, spatial, and basic logic. The expectation is that nobody finishes all questions.
Q. A train leaves station A at 9:00 travelling at 60 km/h. Another leaves station B at 9:30 travelling at 80 km/h in the opposite direction. The stations are 200 km apart. At what time do the trains meet?
By 9:30 the first train has covered 30 km, leaving 170 km between them. Combined closing speed: 140 km/h. Time to cover 170 km: 170 ÷ 140 ≈ 1.21 h ≈ 1h 13m. Meeting time: 10:43.
The trick is not the maths — it's doing the maths in under 30 seconds while resisting the urge to draw diagrams.
Mercer Mettl tests run roughly 30 seconds per question on average. Unlike SHL Verify, you're not expected to finish — the test measures both accuracy and throughput.
Use TestSolve to crack the Mercer Mettl assessment
AI-powered answers and reasoning sent to your phone in seconds. Free first solve, no signup.
Try it free →See pricingThese companies commonly include Mercer Mettl assessments in their hiring process.