Criteria Cognitive Assessment Practice: a free CCAT-style 50-in-15
The Criteria Cognitive Aptitude Test (CCAT) is a 50-question, 15-minute screen: verbal, math and logic, and spatial reasoning mixed together, no calculator, and no penalty for wrong answers. Take a full CCAT-style run below — every question is an original item generated in the same shape — get an instant score with a per-section breakdown, and retake as often as you like. No signup.
An independent practice tool by the QuantVault team · not affiliated with Criteria Corp · how this replica works.
The format you are practicing
The CCAT is Criteria Corp’s flagship pre-employment aptitude test, and a handful of trading firms run it — or a near-identical cousin — as the first gate of their funnel. The run above follows the publicly reported shape:
Firm-specific intel — which Millennium candidates report the screen, where it sits in the funnel, and the technical assessments that follow — lives on the Millennium Criteria assessment page. This page is the format-neutral place to drill reps.
The 18-second budget
- Bank the 5-second points first. An analogy you know and a detail check are 5–10 second items; a two-step percentage is 30. Every question is worth the same point, so harvest the cheap ones at full speed.
- Give a word problem one read. If the setup is not clear after one pass, that is the skip trigger — skip on trigger, not after 40 seconds of struggle. The replica’s skip key (0) exists to train exactly this reflex.
- Never leave a blank. With no penalty for wrong answers, eliminating a single option already makes a guess profitable. When the clock is dying, answer everything you can still reach.
- Do arithmetic the mental-math way. Percentages via 10% blocks, ratios via the unit share, averages via sums. If you want long division, you have misread the question — and if raw arithmetic is the bottleneck, drill the mental-math screens separately.
- Sit full 15-minute runs. Pacing and stamina are half the test. Bursts of ten questions never teach you what minute 13 feels like.
A taxonomy of the spatial items
Spatial questions cost slow candidates the most time per point, and they reduce to two mechanical patterns:
- Next-figure-in-series. The same shape rotates by a fixed step each frame. Do not judge the whole figure: pick one feature — a marked corner, the dot, the stem of the arrow — and track where it must land after one more step. Commit on that single feature.
- Odd one out. Four figures are rotations of one shape; the fifth is its mirror image. Rotation never turns a shape into its reflection, so ask of each figure: could I spin it into the others? The one that needs a flip is the answer.
- Commit fast. Spatial items punish second-guessing more than any other family — your first mechanical read is usually right, and re-checking costs 20 seconds you budgeted for two other questions.
What a good score is (honest numbers)
The only published anchor is the general average: about 24 of 50. Employers do not publish cutoffs, and quant funds certainly do not. What candidate reports suggest: for finance and quant-fund screens, roughly 35+ raw is where reports start calling a run competitive, and the strongest desks are talked about in the low 40s. Treat every one of those numbers as candidate-reported folklore rather than an official bar — and note that speed, not knowledge, is what separates a 30 from a 40. The score card above applies the same bands to your run and tells you which family cost you the most.
Facing it at a specific firm?
Among quant funds, Millennium Management is the name candidates raise most: some report a Criteria assessment as an early screening step, though not consistently across roles and cycles, and the substantive technical OAs are separate and track-specific. Start with the Millennium Criteria assessment replica for the firm-specific reporting, then the Millennium online assessment guide for the market-making take-home and HackerRank tracks that follow. CTC runs the same test format — the CTC aptitude test replica is interchangeable practice — and if your target firm screens with raw arithmetic instead, the mental-math practice hub maps those formats firm by firm.
CCAT practice — FAQ
What is the Criteria Cognitive Assessment (CCAT)?
The CCAT is Criteria Corp's general pre-employment cognitive aptitude test: 50 multiple-choice questions in 15 minutes, no calculator, mixing verbal, math and logic, and spatial-reasoning items. Per public reporting on the format there is no penalty for wrong answers, difficulty ramps up, and most candidates do not finish — the published average raw score is about 24 of 50.
Is this practice test the real CCAT?
No. This is an independent, unofficial practice tool from QuantVault. Every question is an original item written or generated in a similar CCAT-style format — nothing is reproduced from the real CCAT or from any prep provider — and the page is not affiliated with or endorsed by Criteria Corp.
What is a good CCAT score for quant funds?
No employer publishes a cutoff. The published general average is about 24 of 50. Candidate reports put a competitive range for quant-fund screens at roughly 35+ raw, with the strongest desks reported in the low 40s — treat all of these as candidate-reported figures, not official thresholds.
Which quant firms use a Criteria or CCAT-style test?
Millennium Management is the firm quant candidates most often name: some report a Criteria assessment as an early screening step, though not consistently across roles and cycles. CTC runs the same test format. Several mainstream employers outside finance use the CCAT as well.
How should I budget time on a 50-in-15 test?
The average budget is 18 seconds per question. Bank verbal analogies and detail checks in under 10 seconds, cap word problems at about 30, and skip anything you cannot start immediately. With no penalty for wrong answers, always pick something on every question you reach.
Is this CCAT-style practice free?
Yes. The full 50-question, 15-minute run on this page is free and needs no account. Retakes draw a fresh mix from a 120-item hand-checked verbal bank plus parametric math, logic and spatial generators, so no two runs are the same.
How this replica works
Every run is assembled fresh in your browser from a bank of 120 hand-checked verbal items plus parametric generators for word problems, number series, syllogisms, orderings, detail checks and rotated-shape figures — so retakes never repeat a fixed paper, and the selection and order are randomized each time. Each question’s correct answer is re-derived by an independent check in code before it is shown, so the option marked correct always is.
Independence: this page is an independent, unofficial practice tool from QuantVault. It is not affiliated with, sponsored by, or endorsed by Criteria Corp or Millennium Management. CCAT is a trademark of its owner; we use the name to identify the test format being practiced. No real CCAT items — and no prep provider’s items — are reproduced here: every question is an original item written or generated in a similar format. Sat the real screen recently and something feels off? Tell us at [email protected].