McKinsey Solve Game & PST: Complete Preparation Guide

McKinsey’s digital assessment eliminates roughly 70% of candidates before they ever reach a case interview. Whether you face the traditional Problem Solving Test (PST) or the gamified Solve assessment, your preparation strategy determines whether you advance to the interview rounds.

This guide provides everything you need to know about both assessments: what they test, how they’re scored, detailed preparation strategies, week-by-week timelines, and the common pitfalls that trip up even strong candidates.

PST vs. Solve: Understanding the Assessment Landscape

McKinsey has been transitioning from the paper-based PST to the digital Solve game since 2020. While the Solve game is now the default for most offices globally, some regional offices—particularly in parts of Asia and the Middle East—may still administer the PST. Based on our analysis of candidate experiences across multiple recruiting cycles, here’s the current comparison:

The PST passing threshold hovers around 70%, with approximately 30-35% of candidates advancing. Solve uses algorithmic scoring that tracks your mouse movements, click patterns, decision sequences, and time allocation—guessing correct answers won’t help if your approach lacks logical structure.

Why McKinsey Switched to the Solve Game

The traditional PST had several limitations that McKinsey sought to address:

1. Test prep vulnerability: Candidates could memorize question patterns and pass without genuine problem-solving ability
2. Cultural bias: Reading-heavy questions disadvantaged non-native English speakers
3. Limited signal: Multiple-choice format only captured final answers, not thinking process
4. Cheating risk: Paper-based tests in supervised settings were logistically expensive and still vulnerable

The Solve game addresses these issues by measuring how you think, not just what you conclude. The gamified format also creates a more engaging candidate experience and provides McKinsey with richer behavioral data.

The Solve Game: What It Actually Tests

Solve currently features several potential mini-games, with candidates typically facing two per assessment session. The specific games you encounter are assigned algorithmically and may vary.

flowchart TD
    A[McKinsey Solve Assessment] --> B{Mini-Game Assignment}
    B --> C[Ecosystem Building]
    B --> D[Redrock Study]
    B --> E[Sea Wolf]
    B --> F[Plant Defense]
    
    C --> G[Constrained Optimization]
    D --> H[Data Interpretation]
    E --> I[Systems Thinking]
    F --> G
    
    G --> J[Process Score]
    H --> J
    I --> J
    
    J --> K[Final Assessment Result]

Ecosystem Building (Most Common)

The Challenge: You’re presented with a terrain (mountain, coral reef, or similar environment) and given a list of 30-40 species. Your task is to select exactly 8 species that can form a sustainable food chain within the given environmental constraints (altitude, temperature, terrain type, moisture level).

What’s Really Being Tested: Constrained optimization and systematic hypothesis testing. The algorithm watches whether you:

• Read and understand all constraints before starting
• Approach the problem systematically (e.g., starting from producers)
• Adjust your strategy when a selection doesn’t work
• Backtrack efficiently rather than starting over completely

Detailed Strategy:

1. Map the constraints first (2-3 minutes): Before selecting any species, read every species card and note which environmental conditions each requires. Group them mentally by trophic level (producers → primary consumers → secondary consumers → apex predators).

2. Start from one end of the food chain: Working from producers (plants/algae) upward is typically easier because producers have fewer dependencies. Select 2-3 producers that match the terrain, then find primary consumers that eat those specific producers AND match the environmental constraints.

3. Use paper: Create a simple grid tracking each species against the 4-5 environmental constraints. This prevents repeated checking of the same information.

4. When stuck, backtrack one level: If you can’t find a valid apex predator, don’t restart. Replace one secondary consumer with an alternative and check if that opens new apex predator options.

5. Verify the complete chain: Before submitting, trace the energy flow from bottom to top. Every species must have at least one food source below it (except producers) and must satisfy all environmental constraints.

Common Ecosystem Building Mistakes:

• Starting with the most “interesting” species rather than strategically
• Not reading ALL constraints before beginning
• Restarting from scratch after one failed attempt (wastes time and signals poor process)
• Ignoring mutual exclusion rules (some species compete for the same niche)

Redrock Study (Data Interpretation)

The Challenge: You’re given a research packet containing 5-7 pages of ecological research data—charts, tables, scatter plots, and text descriptions. After a reading period, the source material may become partially or fully inaccessible, and you must answer 5-6 questions about the data.

What’s Really Being Tested: Your ability to extract, synthesize, and apply quantitative information under pressure—skills directly applicable to profitability cases and financial analysis.

Detailed Strategy:

1. During the reading phase (critical): Take structured notes. Don’t try to memorize—write down key numbers, trends, and relationships. Organize notes by data source (Chart A, Table B, etc.).

2. Note the scale and units: Many wrong answers come from misreading axes (thousands vs. millions, monthly vs. annual). Write down units explicitly.

3. Identify relationships: Which variables correlate? Which tables connect to which charts? Draw arrows between related data points in your notes.

4. During the question phase: Read each question twice. Identify exactly which data source contains the answer. If you noted it properly, you won’t need to search.

5. Beware of inference questions: Some questions ask what the data proves vs. what it merely suggests. The correct answer must be directly supported by the data, not just plausible.

Common Redrock Mistakes:

• Spending too long on one data source during reading phase
• Not noting units and scales
• Confusing correlation with causation in answers
• Running out of time because notes were disorganized

Sea Wolf (Systems Thinking)

The Challenge: You manage a marine ecosystem or wolf pack, making decisions about resource allocation, territory, and survival strategies across multiple rounds. Conditions change each round (weather, food availability, competitor behavior), requiring you to adapt.

What’s Really Being Tested: Dynamic decision-making and systems thinking. Can you identify which variables matter most, predict second-order effects, and adapt when conditions shift?

Detailed Strategy:

1. Identify the feedback loops: Early rounds often establish baseline relationships. Pay attention to what happens when you change one variable—does it trigger cascade effects?

2. Don’t over-optimize for one metric: The game typically tracks multiple success criteria (survival, growth, territory). Maximizing one at the expense of others usually leads to system collapse.

3. Conserve resources in uncertain conditions: When the environment shifts dramatically, making smaller moves lets you observe effects before committing.

4. Look for leading indicators: Some environmental changes signal upcoming shifts. Early adaptation scores better than reactive scrambling.

Plant Defense (Constrained Optimization)

The Challenge: Similar to a tower defense game, you must protect a habitat by strategically placing defensive plants/organisms. Each placement costs resources, and threats come from specific directions with varying strengths.

What’s Really Being Tested: Resource allocation under constraints, strategic planning, and cost-benefit analysis.

Detailed Strategy:

1. Assess all threats before placing anything: Understand the full scope of challenges before committing resources.
2. Identify chokepoints: Place defenses where they can address multiple threats simultaneously.
3. Reserve resources: Don’t spend everything immediately. Later rounds often introduce stronger threats.
4. Adapt placement based on feedback: If a defense isn’t working, relocate rather than adding more resources to a losing position.

The Five Skills McKinsey Tests (Scoring Criteria)

Understanding the official scoring dimensions helps you practice deliberately. McKinsey’s assessment measures these five cognitive abilities:

How the Algorithm Scores You

While McKinsey doesn’t publish exact scoring formulas, analysis of candidate outcomes reveals several patterns:

Process over outcomes: Two candidates can achieve the same final answer but receive different scores. The one who arrived at the answer through systematic exploration scores higher than one who stumbled upon it through random trial and error.

Time allocation matters: Spending proportionally more time on complex decisions (where more variables interact) and less time on simple decisions signals good judgment about problem complexity.

Consistency is rewarded: Using a clear, repeatable methodology throughout the game scores better than switching approaches mid-stream without cause.

Recovery behavior counts: Making an error isn’t fatal. Recognizing the error quickly, understanding why it happened, and adjusting systematically shows the metacognition McKinsey values.

Estimated Scoring Weights

Based on publicly available candidate feedback and third-party analysis:

What We Know About Passing Thresholds

McKinsey does not publish a specific passing score for the Solve game. However, based on aggregated candidate reports:

• The assessment produces a score on a standardized scale (likely percentile-based)
• Different offices may have different thresholds based on candidate pool size
• Scores are likely compared against other candidates in the same cohort
• There is no “partial pass”—you either advance to interviews or don’t
• Some candidates report receiving feedback that they “demonstrated strong analytical skills” even when not advancing, suggesting the threshold is competitive rather than absolute

For the PST, the passing score is approximately 70% correct (roughly 18-19 out of 26 questions), though this can vary by office and applicant pool competitiveness.

PST Question Types: Detailed Breakdown

If your target office still uses the PST, here’s a comprehensive breakdown of each question type with strategies:

Reading Facts Questions (38% of test)

These are your highest-value preparation target. You’re given a passage or chart and must identify which statement is directly supported by the information provided.

Key techniques:

• Read the question FIRST, then scan the passage for relevant data
• The correct answer is always directly stated or calculable—never requires inference
• Eliminate answers that add information not present in the passage
• Watch for qualifiers: “always,” “never,” “most” can make otherwise-correct statements wrong

Fact-based Conclusion Questions (14% of test)

The hardest question type. You must determine which conclusion is proven (not just suggested) by the combined data from multiple exhibits.

Key techniques:

• The correct answer must be logically inevitable given the data
• “Could be true” is not the same as “must be true”
• Check every exhibit—the answer often requires combining information from 2-3 sources
• Eliminate any answer that requires an assumption not stated in the data

Word Problems (12% of test)

Business math scenarios requiring multi-step calculation without a calculator.

Key techniques:

• Write out the problem structure before calculating
• Round strategically—if answers differ by more than 10%, exact calculation isn’t needed
• Check units at every step (converting monthly to annual, per-unit to total, etc.)
• Work backward from answer choices when direct calculation is complex

Four-Week Preparation Timeline

Based on coaching hundreds of candidates through both assessments, we recommend a structured 4-week preparation approach for candidates with full-time jobs. If you’re a student with more time, you can compress this into 2-3 weeks.

flowchart LR
    W1[Week 1: Foundation] --> W2[Week 2: Skill Building]
    W2 --> W3[Week 3: Simulation]
    W3 --> W4[Week 4: Peak & Rest]
    
    W1 --- W1D[Understand formats\nDiagnostic assessment\nIdentify weak areas]
    W2 --- W2D[Mental math drills\nData interpretation\nStrategy development]
    W3 --- W3D[Full timed simulations\n2-3 per day\nVideo review of process]
    W4 --- W4D[Light practice\nStrategy refinement\nRest before test day]

Week 1: Foundation and Diagnosis (Days 1-7)

Goal: Understand exactly what you’re facing and identify your baseline.

Key deliverable: Know your current level and have a clear skill-building plan.

Week 2: Core Skill Building (Days 8-14)

Goal: Build the fundamental skills tested by both assessments.

Key deliverable: Mental math speed improved by 30%+; can interpret charts in under 60 seconds.

Week 3: Full Simulations (Days 15-21)

Goal: Practice under realistic conditions with full timing.

Key deliverable: Consistent approach across simulations; finishing within time limits.

Week 4: Peak Performance and Rest (Days 22-28)

Goal: Fine-tune and arrive at the assessment fresh and confident.

Key deliverable: Well-rested, confident, with an internalized strategy.

Mental Math: The Foundation Skill

Both PST and Solve require quick calculations without a calculator. In our experience, candidates who invest in mental math see the highest score improvements—often the difference between passing and failing.

Essential Mental Math Techniques

Percentage calculations (most common):

• 10% = move decimal one place left: 10% of 840 = 84
• 5% = half of 10%: 5% of 840 = 42
• 15% = 10% + 5%: 15% of 840 = 84 + 42 = 126
• 1% = move decimal two places: 1% of 840 = 8.4
• Combine: 23% of 840 = 20% + 3% = 168 + 25.2 = 193.2

Division shortcuts:

• Divide by 5 = multiply by 2 and divide by 10: 840/5 = 1680/10 = 168
• Divide by 8 = divide by 2 three times: 840/8 = 420/4 = 210/2 = 105
• For rough division: 2,350/7 ≈ 2,100/7 + 250/7 = 300 + 36 ≈ 336

Compound growth estimation:

• Rule of 72: Years to double = 72/growth rate. At 8%, doubles in ~9 years.
• For 5 years at 8%: multiply by 1.08^5 ≈ 1.47 (or estimate: 100 × 1.47 = 147)
• Shortcut: for small rates over few years, approximate as 1 + (rate × years) + small adjustment. 8% × 5 years = 40%, plus ~7% compound effect ≈ 47% growth.

Daily practice routine (15-20 minutes):

1. 10 percentage calculations (increasing difficulty)
2. 5 division problems (2-4 digit numbers)
3. 3 compound growth estimates
4. 2 multi-step business math problems (revenue = price × volume, margin calculations)

Start with our mental math guide for structured drills that build speed progressively.

Specific Tips for Each Mini-Game

Ecosystem Building: 10 Tactical Tips

1. Spend the first 3 minutes reading only—don’t click anything
2. Create a constraint matrix on paper: species down the left, constraints across the top
3. Count required trophic levels before selecting species (usually need 3-4 levels)
4. Start with producers that match the most terrain constraints
5. Check predator-prey compatibility before confirming any selection
6. When stuck, swap one species rather than rebuilding from scratch
7. Track eliminated species so you don’t recheck them
8. Watch for “keystone” species that connect multiple food chain branches
9. Verify your chain is complete before submitting (every consumer has a food source)
10. If you have extra time, check if alternative combinations might be more robust

Redrock Study: 8 Tactical Tips

1. Use the first 30 seconds to count exhibits and scan question topics
2. Create a “data map” noting what information each exhibit contains
3. Write down exact numbers for any data point that seems important
4. Note trend directions with arrows (↑↓→) rather than trying to memorize values
5. Identify outliers in any dataset—these are frequently tested
6. Check axis labels obsessively—they’re designed to confuse
7. For comparison questions, write the values side-by-side rather than holding in memory
8. Budget time: spend 60% on reading/notes, 40% on answering questions

General Solve Game Tips

• Stable internet connection: Use ethernet if possible; lag can hurt your process score
• Quiet environment: Interruptions force you to re-orient, which looks like inconsistency to the algorithm
• Full screen: Minimize distractions; the assessment tracks focus patterns
• Practice the exact interface if demo versions are available
• Don’t refresh the page: This may reset your session or lose progress

Comparison: Old PST vs. New Solve Game

For candidates who might face either assessment (or who want to understand the evolution), here’s a detailed comparison:

Which One Should You Prepare For?

If you haven’t received confirmation of which format your office uses:

1. Check with your recruiter: They should confirm the assessment format
2. Default to Solve preparation: Most offices now use Solve
3. PST skills transfer: Mental math and data interpretation help with both
4. Prepare for both if uncertain: The skill overlap is significant (60-70%)

Common Mistakes and How to Avoid Them

Strategic Mistakes

Technical Mistakes

Mindset Mistakes

Practice Resources

Free Resources

1. McKinsey official practice: McKinsey occasionally provides sample games on their careers page. Check regularly.
2. Ecosystem simulation games: Games involving food chain building help develop the constrained optimization mindset.
3. Mental math apps: Apps that drill percentages, division, and estimation build PST-critical speed.
4. Logic puzzle sites: Constraint satisfaction puzzles (like advanced Sudoku variants) train systematic elimination.
5. Data interpretation practice: Government statistical publications (census data, economic reports) provide real chart-reading practice.

Structured Preparation

• Our mental math consulting guide builds calculation speed systematically
• Practice with McKinsey-style cases to develop the analytical thinking tested in both formats
• Our AI Mock Interview helps build the structured thinking that Solve rewards

What NOT to Waste Time On

• Expensive “Solve game” courses that promise to “crack the code”—the algorithm is updated regularly
• Memorizing PST questions from forums—the question bank rotates
• Speed-clicking practice—the algorithm measures decision quality, not click speed
• Trying to game the system—McKinsey specifically designs against this

Frequently Asked Questions

How long is the McKinsey Solve game?

The Solve game typically takes 60-90 minutes total to complete. You’ll face two mini-games, each lasting approximately 30-45 minutes. There’s a short break between games. Unlike the PST, you cannot skip ahead or go back between the two games—each must be completed sequentially.

What is the passing score for the McKinsey PST?

The PST passing score is approximately 70% (about 18-19 correct answers out of 26). However, this threshold can vary by office and the competitiveness of the applicant pool that cycle. Some highly competitive offices (London, New York) may have effectively higher thresholds due to stronger candidate pools.

Can I retake the McKinsey Solve game if I fail?

Yes, but typically not for 12-24 months, depending on the specific office’s policy. When you reapply, you’ll take the assessment fresh—your previous score isn’t carried over. Use the waiting period productively: develop analytical skills through case practice, strengthen mental math, and gain problem-solving experience through work or academic projects.

Does the Solve game track mouse movements?

Yes. The Solve game uses behavioral analytics that track mouse movements, click patterns, hover duration, decision sequences, and time allocation between activities. This data helps distinguish between candidates who arrive at correct answers through systematic analysis versus those who reach them through random exploration. Moving your mouse deliberately and clicking with purpose (rather than rapidly clicking around) is important.

How should I prepare differently for PST vs. Solve?

For the PST, focus on mental math speed, chart reading efficiency, and the ability to quickly locate specific information in dense text passages. Practice with timed multiple-choice tests.

For Solve, focus on developing systematic problem-solving habits: always start with understanding constraints, work methodically from one end of a problem, document your thinking on paper, and practice recovering gracefully from mistakes. The PST rewards getting correct answers fast; Solve rewards showing a structured thinking process.

The good news: roughly 60-70% of the underlying skills overlap. Strong data interpretation and logical reasoning help in both formats.

What happens if I don’t finish the Solve game?

Incomplete games receive significantly lower scores. Even a suboptimal solution scores better than no solution. If you’re running out of time, shift to making quick but logical decisions rather than abandoning the game. The algorithm can still extract positive process signals from rushed-but-structured decisions. Leaving a game half-done suggests inability to manage time or prioritize—both red flags for a potential consultant.

Is the Solve game the same for all McKinsey offices?

No. Different offices may use different combinations of mini-games, and the specific scenarios within each game type can vary. The underlying skills being tested are consistent globally, but you shouldn’t assume you’ll get the exact same game another candidate reported. Prepare for all possible mini-game types rather than specializing in just one.

Should I use a practice platform or app to prepare?

Practice platforms can be helpful for familiarizing yourself with the type of thinking required, but be cautious. No third-party platform perfectly replicates McKinsey’s Solve game, and the actual algorithm is proprietary. The best preparation combines: (1) understanding what’s being measured, (2) building genuine analytical skills, and (3) practicing under timed conditions. Platforms that claim to replicate the exact Solve game should be viewed skeptically—McKinsey regularly updates their assessment.

How does McKinsey use the Solve score in the overall application?

The Solve game is a screening tool used primarily to narrow the candidate pool before case interviews. It’s typically treated as pass/fail rather than contributing to an ongoing score. Once you advance past the assessment, your case interview performance determines the final outcome. A borderline Solve score won’t help or hurt you in interviews—what matters is whether you clear the threshold.

After the Assessment

McKinsey typically notifies candidates within 1-2 weeks of completing the assessment. Results are usually delivered via email from your recruiter, with limited detail on specific scores.

If You Advance

Congratulations—you’ve cleared one of the most selective screening filters in consulting. Immediately shift focus to case interview preparation:

• Practice with McKinsey-style cases to familiarize yourself with interviewer-led formats
• Develop hypothesis-driven structures for common case types like profitability and market sizing
• Use our AI Mock Interview to practice under realistic conditions
• Review our McKinsey PST preparation guide for additional context on McKinsey’s evaluation philosophy

If You Don’t Pass

This is not a reflection of your intelligence or consulting potential. The assessment has high variance, and many successful consultants failed their first attempt. Your action plan:

1. Wait the required period (12-24 months depending on office)
2. Build genuine analytical skills: Take on data-heavy projects at work; study statistics or economics
3. Practice case interviews anyway: The structured thinking transfers to the assessment
4. Consider other firms: BCG, Bain, and other top firms have different assessment formats where you might perform better
5. Reapply with a stronger application: Improved resume + stronger assessment = better overall candidacy

Key Takeaways

• McKinsey Solve tracks your process, not just answers—systematic approaches outperform lucky guesses
• The PST passing rate is roughly 30-35%; Solve has similar selectivity
• Allocate 4 weeks of structured preparation (compressible to 2 weeks for students)
• Mental math proficiency is foundational for both assessment formats
• Start from one end of any system (top or bottom), never the middle
• Document your hypothesis-testing process on paper during the assessment
• Complete the game even if imperfect—incomplete submissions score worst
• The algorithm detects random behavior and penalizes it
• Practice under realistic conditions (timed, quiet, full-screen)
• Rest before assessment day—fatigue hurts more than last-minute cramming helps

Ready to Prepare?

Build your analytical foundation with McKinsey cases from our case library. These cases develop the structured thinking and quantitative reasoning that both the Solve game and PST reward.

When you’re ready for realistic interview simulation, try our AI Mock Interview to practice structuring problems and delivering recommendations under pressure—the same skills the Solve game measures through its gamified lens.

For a deeper dive into the traditional test format, see our McKinsey PST preparation guide.