CCAT Prep: A 7-Day Plan for the Criteria Cognitive Aptitude Test

The honest answer about CCAT prep is that a week of focused, timed drilling beats a month of casual reading. The CCAT gives you 50 questions in 15 minutes — that is 18 seconds per question with no calculator and no pencil-and-paper unless you bring it yourself. The candidates who clear Crossover's 35-out-of-50 floor or land in the Vista Equity portfolio's 40-plus range are not smarter; they have rehearsed the format until 18 seconds feels normal. This 7-day plan tells you what to drill on each day, how many questions to do, what raw score to target by the weekend, and the recurring mistakes that cost candidates 5 to 8 raw points they could have kept.

Quick takeaways

• The CCAT is 50 questions in 15 minutes across verbal, math/logic, and spatial reasoning. The average raw score is 24 out of 50.
• Crossover documents a 35/50 floor for almost every role. Vista Equity portfolio roles typically cluster at 40-plus. Software engineer roles at IBM and Capital One sit around 32.
• Day 1 is a diagnostic, not real practice. You need a baseline to know where the 5 to 8 fixable points live.
• 70 percent of useful prep time is timed drilling. 20 percent is reviewing the questions you got wrong. 10 percent is rest.
• The single biggest score leak is the "stuck on one question" trap. The penalty for skipping is one wrong answer. The penalty for stalling 90 seconds is six unattempted questions, which is six wrong answers.
• A realistic 7-day arc moves an unprepared candidate from ~20 raw (30th percentile) to ~32 raw (82nd percentile). That is enough for most software engineering, sales, and operations roles.

What the CCAT actually tests, and where the points leak

The CCAT measures speed of reasoning under time pressure across three rough buckets: verbal (analogies, antonyms, sentence completion), math and word problems, and logic/spatial (number series, matrices, shape rotation). The split is roughly 30 percent verbal, 35 percent math/logic, and 35 percent spatial and abstract reasoning, but Criteria Corp does not publish the exact mix and it varies slightly per form.

What candidates miss in their first attempt is that the CCAT is a speed test dressed as an aptitude test. The questions themselves are not unusually hard. A bright candidate can solve 90 percent of them given unlimited time. The reason the average is 24 out of 50 is that finishing all 50 inside 15 minutes is the actual challenge. Two failure modes account for most of the score gap. First, candidates stall on questions they expect to be able to solve, burning 60 to 120 seconds and losing 4 to 8 questions of throughput. Second, candidates do not internalise that on the CCAT, every unattempted question scores zero — there is no penalty for guessing, so a blind guess is strictly better than leaving a question blank.

The deeper structural issue is that the CCAT does not reward thinking; it rewards pattern recognition. Once you have seen 200 number-series questions, the eighth one in a row resolves in 4 seconds, not 25. The drill volume in this plan is calibrated to get you to that recognition speed in seven days, not seventeen.

The chart below shows how the 7-day arc shifts your time per question and your finishing rate across the test.

The 7-day plan, day by day

Each day in this plan has a single focus, a drill volume, and a target raw score by end of day. Treat the volumes as floors, not ceilings, but do not skip the timed component. Untimed practice is comfort food; it does not move the score.

Day 1: Diagnostic and baseline. Take one full 50-question CCAT simulation, timed, on a laptop, no calculator, no phone in the room. Aim to finish, not to score. Most untrained candidates land at 18 to 22 raw on the first try. Mark every question you got wrong AND every question you skipped. That set of 25 to 30 questions is your real syllabus for the week. Do not look at the answer key for the questions you got right; you will learn nothing.

Day 2: Verbal reasoning, 60 questions. Drill three blocks of 20 verbal questions: analogies, antonyms, sentence completion. Time each block at 6 minutes (18 seconds per question). The verbal section is where most candidates can claw back 3 to 4 raw points fastest, because the patterns are limited. Target end-of-day raw: 23 out of 50 on a fresh mock.

Day 3: Math and word problems, 60 questions. Split into 40 word problems and 20 raw math (percentages, ratios, basic algebra). The CCAT does not test calculus or trig. It tests whether you can set up a percent-change problem in 12 seconds. Memorise common conversions (1/8 = 12.5 percent, 1/3 ~ 33 percent, ratios that share a common factor). Target: 26 out of 50.

Day 4: Logic and number series, 60 questions. This is the day with the steepest learning curve. Number series follow a small set of patterns (constant difference, geometric, alternating, second difference). Once you have drilled 60 of them, the eighth one solves in under 5 seconds. Target: 28 out of 50.

Day 5: Spatial and matrix questions, 50 questions. Shape rotation, odd-one-out, matrix completion. The trap on spatial questions is reading them as visual puzzles when they are usually logical puzzles in disguise. Look for the rule, not the picture. Target: 30 out of 50.

Day 6: Full timed simulation. One complete 50-question test under real conditions: laptop, 15 minutes hard, no breaks, no pause. Then a full review of every wrong answer and every skipped question. Do not take a second mock on Day 6 — review is where the score comes from. Target: 32 out of 50 (82nd percentile).

Day 7: Light review and rest. 15 mixed questions, lightly timed, just to keep the reflexes warm. The rest of the day off. Sleep is a scoring tool the night before a real CCAT, not a luxury.

Day-by-day reference table

The table below summarises the drill volume, focus, and target raw score by day. Bookmark it for the week.

What score to target by role

Target setting is the single most under-discussed part of CCAT prep. A candidate aiming for an entry-level operations role at a regional employer needs a different number than a candidate aiming for the Vista Equity portfolio. The Crossover floor is documented at 35 out of 50. Vista Equity portfolio companies typically cluster at 40 and above. Many software engineering roles at IBM, Capital One, and Lambda School sit around 32. The role's stated CCAT cutoff is the floor, not the ceiling; clearing it by 3 to 5 raw points reduces the risk of being filtered out for "low margin above bar" in a competitive cohort.

Where most candidates lose 5 to 8 points

The 5 to 8 fixable points that separate "passed easily" from "barely passed" come from a small set of recurring mistakes that have nothing to do with intelligence. The honest list:

The first mistake is stalling. Candidates spend 60 to 90 seconds on a question they "almost have." Every additional 30 seconds of stall costs roughly one extra unattempted question at the end. The fix is brutal: if you have not solved a question in 25 seconds, guess and move on. Mark it for review only if you are pacing well.

The second mistake is no-guess discipline. The CCAT does not penalise wrong answers, so blank questions are strictly worse than guessed questions. With four options, a blind guess scores 0.25 expected points. Twelve blank questions left at the end is 3 expected points donated.

The third mistake is reading the spatial questions as art class. Spatial questions are logic questions wearing pictures. The rule is almost always something simple like "rotate 90 degrees clockwise and recolour" or "drop one element." Look for the rule first; the picture comes second.

The fourth mistake is over-investing in the first 10 questions. The first questions are not weighted higher; they are just earlier. Pacing means an even cadence of 18 seconds per question, not "spend 30 seconds on the first ones to feel confident." That is how candidates land at 22 raw with five blanks at the end.

The fifth mistake is misreading the practice landscape. Many free practice tests online are easier than the real CCAT or use a different question mix. The closest free approximations are Criteria Corp's own sample and the question set used in proper paid prep. Build your timing on tests that match the real difficulty curve.

How to drill in the 70/20/10 split

The plan above implies a specific split of where prep time goes. Make this explicit: 70 percent of your week is timed drilling on fresh questions. 20 percent is reviewing the questions you got wrong, writing a one-sentence explanation of the rule you missed. 10 percent is rest and sleep, especially the night before the real test.

Reviewing is what most candidates skip, and it is the most leveraged hour of the week. When you get a number-series question wrong, the lesson is not "I will try harder next time." The lesson is the specific rule you missed: "this series alternates +2 and *3," or "this series has a second-difference of constant 4." Write the rule down. Re-attempt the question 24 hours later cold. That is how the recognition speed compounds.

Timed drilling at random — answering 200 questions across a Saturday with no review — yields almost no score improvement, because you are practicing the same mistakes faster. The candidates who jump 10 raw points in a week are the ones who do 60 questions and review them carefully, not 200 questions and review them carelessly.

FAQ

Can you really raise your CCAT score in 7 days?

Yes, for most candidates. A 7-day plan moves an unprepared candidate from ~20 raw (30th percentile) to ~32 raw (82nd percentile). That is enough for software engineering, sales, and operations roles at most employers. Beyond ~35 raw, the marginal day of prep returns less, because you are now competing on speed ceiling rather than pattern recognition.

Is 7 days enough for a Vista Equity portfolio role?

For most candidates targeting 40-plus raw, 7 days is tight. Add a second week of timed drilling (mostly logic and number series) and a Day 6 mock every other day to push above 36 raw. The Vista Equity portfolio cutoff is the 96th percentile range; that is the top of the curve and the marginal points are the most expensive.

How many full-length mocks should I take in a week?

Two. One on Day 1 (diagnostic) and one on Day 6 (final calibration). More than two full mocks per week burns testing fatigue and the marginal value drops fast. The score gains come from focused sectional drilling and from reviewing wrong answers, not from more full mocks.

Should I memorise the math formulas?

You should memorise a small set: percentages (1/8, 1/4, 1/3, common ratios), unit conversions (minutes to hours, miles to feet), and the standard work/rate setup (rate * time = output). You do not need the quadratic formula or anything beyond pre-algebra. The CCAT does not test calculus or trigonometry.

What happens if I run out of time on the CCAT?

Unanswered questions score zero. There is no penalty for wrong answers. In the final 30 seconds, the optimal play is to fill in every remaining bubble with a guess. A blind guess on four options is worth 0.25 expected raw points per question. Twelve guesses at the end is 3 raw points. That is a meaningful percentile shift at the cutoff.

Is the CCAT proctored?

Most CCAT administrations are unproctored at the application stage and then re-verified at on-site or final-round stages, sometimes with a proctored re-take to confirm the score holds. Crossover specifically re-administers a verified test before offer. Score the way you would on a proctored test, because you may have to.

How does the 7-day plan compare to a 3-day cram?

A 3-day cram gets a candidate from 20 raw to about 26 raw — enough for entry-level roles, not enough for Crossover or Vista. The reason is pure question volume: 3 days is not enough exposure for the spatial and logic patterns to compound into 5-second recognition. If you only have 3 days, prioritise Day 1 diagnostic, Day 2 verbal + math (combined), Day 3 logic + full mock.

What is a passing CCAT score?

There is no universal "passing" score. The CCAT pass mark depends on the employer. Crossover documents 35 out of 50 as a floor; many software engineering roles target 32; entry-level operations roles target 24. Aim for the role-specific number plus a 3-to-5 raw point margin to absorb test-day variance.

Related on PrepClubs

• Pillar. The CCAT in 2026: format, score, and how to prepare. The full brand-level overview of the test.
• Deep practice. CCAT practice with the Pass Guarantee. $39 one time. Pass Guarantee.
• Practice walkthrough. CCAT Practice Test 2026: Free Sample with Walkthroughs. Real questions with the solution method laid out.
• Difficulty. How hard is the CCAT, really. The honest answer about where the difficulty actually sits.
• Guide. What is a good cognitive test score? Per-test benchmarks across CCAT, Wonderlic, PI Cognitive, and SHL.

Practice on PrepClubs

Drill the CCAT under real conditions, with a Pass Guarantee.

PrepClubs CCAT practice runs the full 50-question, 15-minute format on a laptop with no calculator, the same constraints you will face on test day. Every question has a worked solution method, not just an answer key. Sectional drills for verbal, math, logic, and spatial let you target the 5 to 8 fixable points without burning a full mock. $39 one time. Pass Guarantee.

Start free CCAT practice