Mercer Mettl MTAR test 2026 guide

Updated April 2026 · 10 min read · Mercer Mettl Talent Assessment for Recruitment

MTAR (Mercer Mettl Talent Assessment for Recruitment) is the configurable cognitive-and-coding battery used across many Indian campus and lateral hiring drives. Mercer Mettl operates as a platform — the employer configures which sections appear, the difficulty level, and the scoring weights. This means MTAR experiences vary significantly between Wipro NTH, ZS Associates, and Capgemini, even though the underlying platform is the same.

MTAR core sections

Quantitative ability

Number system, percentages, ratio-proportion, time-distance-speed, basic algebra, geometry, data interpretation. Difficulty similar to GRE or CAT quant. Tight per-question timing (~60-90 seconds).

Logical reasoning

Coding-decoding, blood relations, syllogisms, seating arrangement, direction sense, statement-conclusion. Standard Indian-exam style logical patterns.

English / verbal ability

Vocabulary, error spotting, reading comprehension, sentence rearrangement. Calibrated to Indian English-medium graduate level.

Coding (Wipro NTH and similar)

For software engineering roles, MTAR adds a coding round: 2-3 problems in 60 minutes, multiple language options (C, C++, Java, Python). Test cases run live; you see immediate pass/fail. Wipro NTH (National Talent Hunt) and similar IT services drives use this variant.

Behavioural / personality

Workplace personality questionnaire. Standard rating-scale format. Less commonly used as a hard cutoff; more often as a tiebreaker.

MTAR variants by employer

Scoring

MTAR scores each section independently then aggregates with employer-defined weights. Cutoffs are often per-section (e.g., 60% in Quant, 60% in Logical, plus aggregate 65%) rather than purely aggregate. This means a low score in any single section can disqualify even with strong other sections.

Preparation strategy

Per-section discipline. Because MTAR uses per-section cutoffs, you can't trade off. Allocate equal practice time across Quant, Logical, and English even if one is your strength.

Quant fundamentals. R.S. Aggarwal's "Quantitative Aptitude" or M.K. Pandey's "Analytical Reasoning" cover ~80% of MTAR question patterns. Drill these.

Coding round preparation. If your variant includes coding, practice on Mercer Mettl's official sample tests rather than generic LeetCode. The IDE behaviour is specific and the test cases are more straightforward than LeetCode hard problems — clarity beats cleverness.

Time discipline. MTAR sections are timed independently. Don't carry over time from one section to another.

Understanding which MTAR variant you're facing

Because MTAR is a configurable platform rather than one fixed test, the single most useful preparation step before you begin studying is confirming exactly which sections your specific drive includes — a Wipro Elite NTH candidate preparing only Quant and Logical while a coding round is actually included in their drive will waste valuable prep time on the wrong balance. Check the official communication from the employer or campus placement cell carefully; it typically lists the section breakdown, though the terminology can vary (some communications say "aptitude round" to mean Quant + Logical + English combined, others break each out separately). If your communication is genuinely ambiguous, checking recent candidate reports for the same employer's current hiring cycle — rather than relying on a prior year's format, since these can change — is more reliable than assuming a generic MTAR structure applies.

A closer look at the coding round

For roles that include it, the coding round is often the section candidates prepare for least effectively, since general competitive-programming practice (LeetCode-style hard problems) doesn't map cleanly onto what MTAR's coding round actually tests. MTAR's problems tend to be more straightforward algorithmically — array manipulation, string processing, basic data structure usage — but score heavily on getting a fully working solution that passes all visible and hidden test cases within the time limit, rather than rewarding a technically elegant but incomplete solution. Since test cases run live and give immediate pass/fail feedback, a practical strategy is to get a working, if inelegant, solution passing as many test cases as possible early, then spend remaining time refining it — rather than spending the first 20 minutes designing an optimal approach on paper before writing any code, which risks leaving too little time to actually implement and debug it.

The English/verbal section, in more depth

Because this section is calibrated to Indian English-medium graduate level rather than a general international audience, some Indian test-prep-specific patterns are worth knowing beyond general vocabulary building: error-spotting items typically embed exactly one grammatical error per sentence (a subject-verb agreement slip, a wrong preposition, an incorrect tense) rather than multiple errors, so once you've found one plausible error, resist the urge to keep hunting for a second. Sentence-rearrangement items usually have one clear "anchor" sentence — often containing a pronoun or connective word that only makes sense following a specific other sentence — and identifying that anchor first, rather than trying to order all sentences from scratch, is generally the faster approach under time pressure.

Logical reasoning: which sub-types show up most

Of the logical reasoning sub-types MTAR draws from, seating arrangement and syllogisms tend to consume the most time relative to how many points they're worth, since both require building an internal model (a seating chart, a set of logical relationships) before you can answer any question in that item's cluster — meaning the first question in a seating-arrangement cluster costs disproportionately more time than the follow-up questions once the arrangement is worked out. Blood relations and direction sense items are typically faster once the underlying notation (family-tree shorthand, compass-direction tracking) is familiar, since each item is usually self-contained rather than building on a shared setup. A sensible time-allocation habit: when you hit a seating-arrangement or syllogism cluster, budget slightly more time for the first question in that cluster specifically, since the follow-up questions in the same cluster typically go much faster once the initial model is built, freeing up genuine extra time elsewhere in the section, which you can then redirect toward whichever specific question types you personally find hardest to work through quickly, accurately, and confidently under real time pressure on test day itself.