Concepts
What each category measures, and what your result means.
Deductive Reasoning
What it measures
Measures classic deduction — whether a conclusion is strictly guaranteed by the premises you're given, not just consistent with them. Expect syllogisms, knights-and-knaves truth/lie puzzles, and logic-grid problems with several entities and attributes that you narrow down by elimination.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Because deductive-reasoning tests order their questions from easiest to hardest, the difficulty breakdown on your report shows roughly where a chain of premises got too long to hold in your head — a better place to focus practice than the percentage alone.
Numerical Sequences
What it measures
Each item in this category hands you a short run of numbers and asks you to continue it — the challenge is reverse-engineering the pattern generator from only a few outputs rather than following an established formula. Some sequences settle into a steady arithmetic climb, others accelerate through second-order differences, run on ratios, or braid two separate patterns into one alternating list.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. The difficulty breakdown shows where the underlying rule stopped being a simple constant difference and started requiring ratios or second differences — that's usually the spot worth another look.
Figural Series
What it measures
A strip of panels shows one figure evolving frame by frame, and your job is to continue the sequence purely by eye — no words involved anywhere in the reasoning. The shape might spin, gain or lose fill, grow, shrink, or multiply between frames, so the real skill is isolating exactly which visual property is doing the changing before guessing what comes next.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Because these tests get harder as they go, a lower band usually means the transformation rule stopped being a single visual property and started combining two or three at once — check the explanations for the exact rule you missed.
Verbal Reasoning
What it measures
This category puts a passage in front of you and asks you to separate what it locks down from what it only gestures at, alongside items where you name the specific flaw in an argument's logic and others where you work out the relationship linking a pair of words well enough to complete a second pair.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. If your band is lower than you expected, the explanations are worth reading closely — this category's most common trap is treating 'sounds plausible' as if it were 'logically guaranteed,' a distinction that's easy to miss under time pressure.
Spatial Reasoning
What it measures
Every item asks you to picture an object changing in space rather than on paper — turning a flat or solid shape over inside your mind, tracking a sequence of folds, or deciding whether two figures are really the same piece from a different angle or secretly mirror images of each other.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. This category's classic trap is a mirror image disguised as a rotation, so if your band is lower than expected, check whether the questions you missed cluster around that specific confusion before assuming the whole skill needs work.
Truth Tables & Syllogisms
What it measures
You'll work through compound logical statements the way a computer would — plugging in true/false values for each piece and tracking how AND, OR, NOT, XOR, and IF-THEN combine them into a final result, with tools like De Morgan's Laws available to shortcut the brute-force row-by-row approach.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Because these problems build outward from the innermost parentheses, the difficulty breakdown tends to show whether you're solid on single connectives but start slipping once statements nest two or three levels deep.
Raven-Style Patterns
What it measures
A grid of related figures is missing its final cell, and you have to figure out which candidate belongs there — which means decoding a pattern that has to hold simultaneously in the row and in the column, not just one direction at a time, with no words or numbers involved.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. A lower band here usually means you found a rule that fit part of the grid but not all of it — the explanations lay out the full row-and-column rule you'd need to check every candidate against.
Estimation Reasoning
What it measures
You're handed a quantity nobody could just look up — how many piano tuners work in a city, say — and asked to build a rough answer out of smaller guesses chained together. Precision on any single step doesn't matter; what matters is landing in the right general range, the classic 'Fermi problem' skill that consulting interviews are famous for testing.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Since these questions accept a range of reasonable answers rather than one exact number, a lower band usually points to the estimation chain itself — check the solutions for which intermediate factor you rounded too far off, rather than assuming you needed exact figures you didn't have.
Data Interpretation
What it measures
Numbers arrive packaged in tables, graphs, and diagrams here, and the task is pulling the right figures out and combining them correctly — while staying alert to visuals built with a truncated axis or a skewed scale that make the underlying numbers look more dramatic than they actually are.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. This category's most common miss is misreading what's actually being measured, so a lower band is worth revisiting with the table or chart still in view — check units and axis labels before the calculation itself.
Probability Reasoning
What it measures
Counting problems, conditional-probability puzzles, and questions built around rare events all live in this category, and a lot of them are designed to tempt an intuitive but wrong answer — most often by burying a low base rate under a vivid, distracting detail that feels more relevant than it is.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. If your band is lower than expected, look for a pattern in which questions you missed — base-rate problems and combinatorics call for genuinely different techniques, so a weak spot in one doesn't necessarily mean the other needs work too.
Mechanical Reasoning
What it measures
Each question shows a small mechanical setup — a lever, a pair of meshed gears, a pulley arrangement, or a sealed tube of fluid — and asks what happens somewhere else in the system when one part moves or a force is applied, testing the kind of physical intuition you'd build from tinkering rather than from a textbook.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Since each question usually hinges on one physical principle, the explanations are the fastest way to see whether a lower band traces back to gear ratios, lever balance, or fluid pressure specifically — worth pinning down before your next attempt.
Quantitative Aptitude
What it measures
Most questions here are everyday math dressed up as a story problem — splitting a bill by ratio, working out interest or a discount, optimizing a route or a rate — with one twist: a subset of 'data sufficiency' questions ask you to decide whether the numbers you've been given are even enough to solve the problem, without requiring you to grind through the actual solution.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Data-sufficiency items and standard word problems test different instincts, so if your band is lower than expected, the explanations will show whether the miss was arithmetic or a case of combining two pieces of information before checking each one alone.
Attention to Detail
What it measures
You're shown pairs of codes, strings, or images that look identical at a glance, and the job is catching the one character or pixel that doesn't match before the clock runs out — a test of careful scanning under pressure more than of any deeper reasoning skill.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. Because this category rewards a consistent scanning method more than raw speed, a lower band is usually fixed by slowing down on a re-check pass rather than trying to go faster on the first one.
Conditional Reasoning
What it measures
This category zeroes in on a single logical form — 'if P, then Q' — and tests how reliably you can operate on it: recognizing when a condition is required versus merely enough on its own, linking several conditionals into a valid chain, and catching the classic trap of treating a rule's converse or its inverse as if it followed just as surely as the contrapositive does.
What your result means
Your score is the percentage of questions you answered correctly, mapped onto five qualitative bands — Excellent, Very good, Good, Developing, and Needs practice — with unanswered questions counted as incorrect. This category's single most common trap is mistaking the converse or inverse of a rule for something logically guaranteed, so if your band is lower than expected, check whether your misses cluster around that exact confusion before assuming the whole format is the problem.
Cognitive Ability
What it measures
This one session throws five different puzzle types at you back to back — a matrix to complete, a number series to extend, a verbal analogy to solve, a logical deduction to work through, and a spatial figure to rotate — because combining several reasoning formats into a single timed sitting is exactly how real hiring platforms operate too, TestGorilla, the CCAT test from Criteria, and SHL's broader aptitude battery among them. With 25 questions to get through in 25 minutes and no way to revisit an earlier answer, part of what's being measured is how fast you can reorient when the puzzle type changes under you.
What your result means
Your result here works differently from every other category: instead of a simple percentage, each question is weighted by its difficulty level (1 to 5), and your normalized score is the weighted sum of your correct answers over the maximum possible, expressed on a 0-100 scale and placed into one of four bands — Developing, Solid, Strong, or Exceptional. This score is an internal practice metric only: it is not a certified psychometric instrument and does not measure your real IQ, so treat the by-area breakdown (matrix reasoning, number series, verbal analogies, logical deduction, spatial visualization) as a map of where to practice next, not a diagnosis.