Aptitude Tests for Consulting: Every Firm's Test Explained (2026)

Consulting aptitude tests are timed online assessments that screen candidates before live case interviews. In 2026, every major consulting firm uses one — McKinsey's Solve, BCG's Casey plus the Consulting Career Assessment (CCA), Bain's SOVA or Pymetrics, plus dozens of tier-2 and Big 4 variants built on SHL, Cappfinity, and Kenexa. This guide is the full map: what each firm uses, how it's scored, and a 3-week prep plan that works for all of them.

From Road to Offer's aptitude test practice sessions, the single highest-leverage prep activity isn't more practice questions — it's learning the specific test's scoring logic. For McKinsey Solve, the process score rewards systems thinking over math speed. For BCG Casey, it's factor prioritization over exhaustive analysis. For SOVA, it's speed-accuracy tradeoff. Most candidates prep as if every test rewards the same thing. Here's the scoring truth for each firm.

The Full Map: Every MBB, Tier-2, and Big 4 Aptitude Test

Three patterns matter. First, MBB firms run proprietary game-based tests (Solve, Casey, Pymetrics) on top of standard psychometric screens. Second, tier-2 and Big 4 firms lean on SHL, Cappfinity, and Kenexa — meaning one solid SHL prep set covers half the tier-2 market. Third, regional variation is real: Bain US typically runs HireVue or TestGorilla, Bain EMEA runs SOVA, Bain Germany runs Pymetrics.

McKinsey Solve: What It Measures

McKinsey Solve is a ~65-minute assessment with two games — Redrock Study (~35 min) and Sea Wolf (~30 min) — plus an 85-minute variant in some offices that adds the Sustainable Futures Lab (Source: CaseBasix, McKinsey Solve Guide 2026). The Ecosystem Building game was retired globally in July 2025.

Solve measures three things. Data interpretation dominates Redrock — six mini-cases on a wolf/elk ecosystem with tables, charts, and driver trees. Optimization dominates Sea Wolf — a 30-minute puzzle where you build a marine food web under energy and predator constraints. The process score tracks how you navigate: reading time, answer revision patterns, and whether you click through systematically or randomly.

Scoring truth: The product score (right answers) and process score (behavioral signals) both count. Random clicking to finish fast scores worse than reaching fewer correct answers through systematic reading. McKinsey publishes no thresholds, but PrepLounge and Wall Street Oasis data suggest ~20-30% of takers advance to interviews. See the full McKinsey Solve guide, Sea Wolf strategy, and Redrock study guide for module-by-module drills.

BCG Casey and CCA: What They Measure

BCG runs two assessments, in sequence. The Consulting Career Assessment (CCA) comes first — a 30-minute SHL-built battery with 76 personality prompts, 8 numerical reasoning questions (12 minutes), and 15 more personality prompts (Source: Prepmatter, BCG CCA Guide 2026). Introduced in August 2024, the CCA screens for behavioral fit before any case work.

After passing the CCA, candidates take the BCG Casey chatbot case — an interactive case simulator with 8-10 questions plus a 1-minute video recommendation (Source: CaseBasix, BCG Casey Chatbot Guide 2026). Total Casey time: ~30-35 minutes. You cannot revisit submitted answers, and there is no pause button.

Scoring truth: Casey rewards factor prioritization over exhaustive analysis. When the chatbot asks "which factors matter most?", listing everything relevant scores worse than picking the three drivers that actually move the decision. Video recommendations are evaluated on structure (MECE, CEO-ready) and confidence, not polish. Read the full BCG Casey guide for question-by-question drills.

Bain SOVA: What It Measures

Bain's SOVA test is a 5-section, 60-75 minute assessment totaling ~75 questions (Source: Hacking the Case Interview, Bain SOVA Test 2026). The sections: Situational Judgment (workplace scenarios, most/least effective responses), Verbal Reasoning (True/False/Cannot Say), Numerical Reasoning (chart and table interpretation), Logical Reasoning (pattern and sequence), and Personality (1-5 self-assessment scale).

Scoring truth: Dual scoring — accuracy is primary, speed differentiates. Two candidates who each get 12/15 numerical questions right score differently if one took 10 minutes and the other 18. No visible timer appears, but response time per question is tracked. Cross-section consistency matters: a strong numerical score does not compensate for a weak verbal score. SOVA benchmarks against consulting-specific norms, not the general population. Read the full Bain SOVA guide for section-by-section pacing.

Kearney, L.E.K., and Oliver Wyman: Tier-2 Variants

Tier-2 firms run shorter, math-heavier assessments. The Kearney test combines numerical reasoning with a structured written case component (~45-60 minutes). L.E.K. runs a numerical + verbal battery (~40-55 minutes) with heavy business-math weighting reflecting the firm's life sciences focus. Oliver Wyman's pre-interview screening varies by office but typically includes mental math and exhibit interpretation in the ~30-45 minute range, reflecting the firm's candidate-led case style.

Strategy& and EY-Parthenon use SHL-built batteries (numerical + verbal + deductive + SJT) that run 60-90 minutes total. Cappfinity powers EY-Parthenon's strengths and numerical modules — the Cappfinity numerical test has ~20 questions with no hard time limit but speed-and-accuracy scoring (Source: GraduatesFirst, Cappfinity Assessments 2026).

The Generic Layer: SHL, Kenexa, Cappfinity Numerical Reasoning

Most Big 4 consulting tests are built on SHL, Kenexa, or Cappfinity numerical reasoning engines. Understanding the generic format covers 60%+ of tier-2 and Big 4 tests.

SHL numerical reasoning: 18-20 questions in 20-25 minutes. Multiple choice. Each question presents a chart, table, or graph with 3-5 questions attached. Answer level: 3-4 layers of calculation. Speed-and-accuracy scored.

Kenexa numerical reasoning: 20-30 questions in 20-30 minutes for general roles, ~32 questions in 35 minutes for graduate or managerial roles (Source: GraduatesFirst, IBM Kenexa Assessment Tests 2026). Multiple choice.

Cappfinity numerical reasoning: ~20 questions with three formats — multiple choice, ranked ordering, and free-form calculation. No hard time limit but measured on speed and accuracy (Source: GraduatesFirst, Cappfinity Assessments 2026).

Worked Example: A Sea Wolf-Style Ecosystem Question

Prompt: You are given 40 marine species. Build a food web that supports the maximum total biomass of apex predators, subject to: (1) each predator needs at least 3 prey species, (2) total energy input from sunlight is capped at 10,000 kcal/day, (3) no prey species can appear in more than 2 predators' diets.

Step 1 — Rank candidates by biomass-per-energy ratio. Pick the 4 apex predators with the highest biomass per kcal of prey consumed. Skip the two heaviest predators if their energy requirement blows the 10,000 kcal cap.

Step 2 — Assign prey under the 2-predator cap. Start with the rarest prey species. Assign each to the predators who benefit most per kcal. Track the cap: once a prey species hits 2 predators, lock it out.

Step 3 — Check constraint compliance. Every predator has ≥3 prey. Total sunlight energy ≤ 10,000 kcal. No prey species in >2 diets. If any constraint fails, swap the lowest-biomass predator for the next candidate.

Step 4 — Validate total biomass. Sum apex predator biomass. This is your product score. The process score tracks whether you read constraints before clicking, revisited answers systematically, or clicked randomly — systematic navigation adds to the final evaluation.

Why this matters: Sea Wolf's penalty is 20% deduction for each missed constraint. Candidates who rush and miss the 3-prey minimum for one predator lose a full grade band. The test rewards reading every constraint before moving, not speed-clicking through.

Test Your Aptitude Test Knowledge

How to Prep: A 3-Week Plan for Any Consulting Aptitude Test

Week 1 — Generic foundations (covers 60% of tests). Six 30-minute sessions on SHL-style numerical reasoning (chart/table interpretation), verbal reasoning (True/False/Cannot Say), and mental math shortcuts. Free sets on JobTestPrep, PracticeAptitudeTests, and Road to Offer's math drills.

Week 2 — Firm-specific modules. Drop your generic SHL reps and spend the week on your target firm's actual test. McKinsey → Solve simulations (CaseBasix free simulation, MConsultingPrep drills). BCG → Casey chatbot practice + CCA sample. Bain → SOVA practice set + Pymetrics walkthrough. Do 3 full run-throughs of each module before test day.

Week 3 — Simulation and scoring review. One full-length mock per day, timed, with no interruptions. After each mock, review the wrong answers and categorize: (1) did you misread the question, (2) did you miscalculate, or (3) did you mis-prioritize? Most gains come from fixing pattern 3 — scoring logic, not raw ability.

Sources

Sources checked April 12, 2026:

• CaseBasix, McKinsey Solve Game Guide 2026
• Prepmatter, BCG Consulting Career Assessment (CCA) Guide
• CaseBasix, BCG Online Case (Casey Chatbot) Guide 2026
• Hacking the Case Interview, Bain SOVA Test 2026
• GraduatesFirst, IBM Kenexa Assessment Tests 2026
• GraduatesFirst, Cappfinity Assessments 2026
• PrepLounge, Aptitude Tests in the Consulting Industry
• IGotAnOffer, Bain Online Assessment Guide 2026
• StrategyCase, BCG Cognitive Test 2026
• Road to Offer, Consulting Assessment Tests Guide

Related Guides

• McKinsey Solve Complete Guide
• McKinsey Sea Wolf Strategy
• McKinsey Redrock Study Guide
• McKinsey Ecosystem Building Guide
• BCG Online Case Casey Guide
• Bain SOVA Test Guide