Last Updated on April 21, 2026
The McKinsey Sea Wolf game (part of the McKinsey Solve assessment) is a 30-minute optimization simulation where candidates must select microbial treatments under time pressure. It is where most rejected applicants lose their chance at a first-round interview. Underestimating it is the default mistake. After collecting hundreds of test-taker debriefs, I can tell you exactly how the scoring works, which mistakes cost the offer, and how to optimize under the time pressure that breaks most first-timers.
Key Takeaways
- The Sea Wolf game is a 30-minute optimization puzzle with three contaminated sites, not a biology test; the correct answer is the best microbe combination available, not a perfect one.
- You must select exactly two characteristics to filter microbes in Phase 1, then build a 10-microbe prospect pool before committing to a three-microbe treatment.
- McKinsey scores portfolio thinking and speed; the effectiveness ceiling per site is typically 80 to 100%, so chasing perfection burns time you need for Sites 2 and 3.
- Allocate 12 minutes to Site 1 and 9 minutes each to Sites 2 and 3. Practiced candidates finish each site in around 7 minutes.
- Sea Wolf sits between Red Rock Simulation and Sustainable Future Lab, giving a total Solve runtime of 85 minutes.
What Is the McKinsey Solve Sea Wolf Game?
The Sea Wolf game (also called the Ocean Cleanup or Ocean Treatment module) is a current component of McKinsey’s Solve assessment that asks candidates to design microbial treatments for three plastic-contaminated ocean sites. It has been a standardized module since late 2023 and is now the second game every Solve candidate sees.
The mechanics are identical across all three sites. You are given a site-specific treatment profile, a pool of microbes with different attributes and traits, and a fixed set of filters. Your job is to identify the three microbes whose averaged attributes and collective traits best match the site’s requirements.
It looks like a biology problem. It is actually a constrained optimization problem dressed up with environmental storytelling. McKinsey is watching how you filter, how you trade off, and how you manage your 30 minutes, not whether you know anything about marine microbiology.
Where Sea Wolf Sits in the McKinsey Solve Sequence
McKinsey runs three modules in sequence, and the order has been consistent since early 2026:
1. Red Rock Simulation (analytics-focused research report generation)
2. Sea Wolf (the microbe-selection module covered in this guide)
3. Sustainable Future Lab (behavioral and prioritization module added in April 2026)
Total runtime is 85 minutes across the battery. Sea Wolf takes 30 of those minutes.
Before the recent changes, Sea Wolf was sometimes swapped with other production modules. That is no longer the case. If you are sitting the Solve in 2026, you will see Sea Wolf, and it will be Module 2. Plan your preparation around it.
Candidates who want the full sequence (plus the changes McKinsey rolled out in April) can read our McKinsey Solve Game article for the complete picture.
The Objective: What You Are Actually Solving
Every site in Sea Wolf gives you the same task:
“Select 3 microbes whose averaged attributes and collective traits best match the site’s required treatment profile.”
Two things matter in that sentence.
Averaged attributes means each microbe has three numerical values (Density, Energy, Size, each on a 1 to 10 scale), and the average of your chosen three microbes has to fall inside the site’s required range. If the site wants Density between 6 and 8, the average density of your three microbes has to land in that window.
Collective traits means each microbe has traits like Aerobic, Heat Resistant, Hydrophilic, or Light Sensitive. The site tells you which one trait you need (desired) and which one trait you should not have (undesired). At least one of your final three needs the desired trait.
The scoring is weighted. Get the averages wrong and your treatment effectiveness drops fast. Include the undesired trait and you lose credit for that site. Miss the desired trait entirely and you lose partial credit but can still score.
The Four-Phase Sea Wolf Game Flow (Step-by-Step)
Every site runs through the same four phases. Once you know the pattern, Sites 2 and 3 become faster because the mechanics repeat.
Phase 1: Filter Site Characteristics
You start at the site interface. On the top right is the Site Requirements Panel: three required attribute ranges and two traits (one desired, one undesired). On the left is the Characteristics Selection Panel with attribute toggles and trait toggles.
The instruction reads: “Select 2 microbe characteristics according to your current site information.”
You must pick exactly two characteristics. Not three, not one. This is a filter, not a ranking.
The right choice depends on which constraints are tightest for that specific site. If the Density range is narrow (for example, 6 to 8) and the Size range is wide (for example, 2 to 10), filter on Density. If the undesired trait would eliminate many microbes in the pool, filter on that trait. Choose the two filters that will shrink the candidate pool the most while keeping viable options visible.
A common mistake: candidates pick the two most interesting characteristics instead of the two most restrictive. The game rewards filter efficiency, not variety.
Phase 2: Evaluate and Shortlist Microbes
Once your filters are set, you evaluate microbes one by one. Each microbe has its own attribute values and trait set.
Your job in Phase 2 is to:
- Eliminate any microbe that violates a site attribute range
- Exclude every microbe carrying the undesired trait
- Keep at least a few microbes that carry the desired trait
- Preserve enough variety that you can still balance averages in Phase 4
Treat this like a hypothesis-driven screening pass. You are not picking a single winner; you are narrowing a funnel. If you kill too many microbes here, you run out of options in Phase 4 and cannot hit the required averages.
Phase 3: Build the Prospect Pool
Phase 3 is where most candidates lose the game without realizing it.
You start with six microbes already in your prospect pool. The game then runs four selection rounds. In each round, you are offered three candidate microbes and must pick one. Those four selected microbes expand your pool to a total of ten.
Here is the mental model most candidates miss: none of the three options per round will be perfect. The game is designed that way. You are choosing the best available, not the ideal. Each round is a local trade-off in service of a global portfolio.
Think about what your current pool already covers. If you already have three Hydrophilic microbes and the site needs one Hydrophilic, you do not need a fourth. If your average Density is already at the high end of the range, pick the microbe that pulls it down, not one that pushes it further up.
This is portfolio thinking. And it is exactly what McKinsey is testing. The game calls it “optimization under constraints, not idealized selection,” which is a polite way of saying: stop looking for the microbe that fixes everything.
The effectiveness score for each site is capped. In practice, treatments land between 80% and 100%, so a “good enough” portfolio that covers the constraints will outscore a half-built perfect portfolio every time.
Phase 4: Finalize the Three-Microbe Treatment
The last phase is the commitment. From your 10-microbe prospect pool, pick the three microbes that:
- Produce averaged attribute values inside every required range
- Collectively include the desired trait
- Collectively exclude the undesired trait
The most common Phase 4 error is math pressure. Candidates who have not tracked running averages during Phase 3 now face a sudden arithmetic test and panic. If you have been sanity-checking averages through Phases 2 and 3, this step takes two minutes. If you have not, it takes more, and you blow your time budget.
After submitting Site 1, the game resets to a new environmental profile and you repeat the full four-phase sequence for Site 2, then Site 3.
What the Sea Wolf Game Actually Tests
McKinsey has told nobody publicly what Sea Wolf measures. But the design tells us, and 600+ test-taker debriefs confirm it.
Three skills carry the weighting:
1. Structured filtering under ambiguity. Translating a site profile into the right two filters is the single most predictive move. Candidates who pick the wrong filters in Phase 1 rarely recover. This is the same skill a first-year associate uses when a partner drops 200 pages of data on their desk and asks “so what?”
2. Portfolio optimization under constraints. Phase 3 rewards candidates who think in averages, covers, and trade-offs, not in best-in-class picks. Real consulting work is the same: there is no perfect answer, only the best answer given the constraints.
3. Learning curve speed. Sites 2 and 3 use the same mechanics as Site 1. A candidate who needed 12 minutes for Site 1 and still needs 12 minutes for Site 3 has not learned. A candidate who finishes Site 3 in 7 minutes is showing the pattern recognition McKinsey wants in its consultants.
You can see why this module predicts interview performance. It is a microcosm of the work itself.
Time Management for the 30-Minute Sea Wolf Module
The clock is the hidden adversary. Candidates who run out of time on Site 3 score zero on that site, regardless of how well they did on Sites 1 and 2.
Based on our test-taker data, the optimal time split is:
| Site | First-time target | After 3+ practice runs |
|---|---|---|
| Site 1 | 12 minutes | 7 minutes |
| Site 2 | 9 minutes | 7 minutes |
| Site 3 | 9 minutes | 7 minutes |
| Buffer | 0 minutes | 9 minutes |
First-time candidates should front-load Site 1 because that is where the mechanics are learned. By Site 2, the interface is familiar and the same filters-screen-pool-commit sequence applies. By Site 3, a practiced candidate can compress the whole flow.
Failing to move on when stuck is the other killer. If you have spent 11 minutes on Site 1 and the averages still will not balance, commit to your best available three and move on. The marginal gain on Site 1 from a twelfth minute is always less than the marginal gain on Site 3 from having a full nine minutes instead of seven.
Common Mistakes That Cost Candidates the Sea Wolf Game
From reviewing candidate debriefs, the patterns are consistent.
Mistake 1: Filtering on the wrong two characteristics. Candidates default to the first two attributes they see instead of choosing the filters that prune the pool most aggressively. Pick the tightest constraint and the undesired trait whenever possible.
Mistake 2: Hunting for a perfect microbe. Sea Wolf rewards covering all constraints, not nailing any single one. If a microbe fixes one attribute but wrecks another, it is worse than the “average” microbe that balances everything.
Mistake 3: Ignoring running averages. Candidates who do not sanity-check averages during Phase 3 are forced into arithmetic panic in Phase 4. Keep a rolling mental estimate of where your pool sits on each attribute.
Mistake 4: Misreading “at least one desired trait” as “all microbes must have the desired trait.” The instruction is collective, not individual. One of your three final microbes with the desired trait is enough. Do not disqualify microbes that lack it.
Mistake 5: Clock blindness on Site 3. Candidates who treat all three sites as equal and run 10-10-10 usually crash out of Site 3 with incomplete treatments. Budget time backwards from Site 3 and protect the buffer.
Mistake 6: Treating Sea Wolf as a biology test. It is not. Knowing what “aerobic” means in real microbiology is irrelevant. The traits are game labels, not scientific signals.
How to Prepare for the McKinsey Solve Sea Wolf Game
You cannot cram microbe trivia and win. The only preparation that moves your score is repeated exposure to the four-phase structure under time pressure.
The three preparation priorities:
1. Practice the four-phase flow until it is automatic. You want Phases 1 and 2 on autopilot so that your conscious attention during the real test is on Phase 3 trade-offs. Running through 10 to 15 practice sites is usually more than enough to reach that state.
2. Build a quick mental math habit for three-value averages. You will be averaging three numbers on a 1 to 10 scale against a site range. Practiced candidates do this in their head in under 10 seconds. Candidates who count on their fingers burn time they do not have.
3. Learn to stop optimizing. This is the hardest one. Every practice site should end with a conscious decision of “this is good enough, submit.” Training yourself to walk away from marginal perfection is what separates top-quartile Solve scorers from the rest.
Frequently Asked Questions
Is the Sea Wolf game hard?
Yes, harder than it looks. The mechanics are simple but the combination of constrained optimization and time pressure across three sites separates prepared candidates from unprepared ones. Most first-time test takers run out of time on Site 3.
What score do you need to pass the Sea Wolf module?
McKinsey does not publish thresholds, but based on our candidate data, the typical cutoff for advancing to first-round interviews corresponds to roughly 80% treatment effectiveness across all three sites combined. You can miss 20% and still pass.
How long is the McKinsey Sea Wolf game?
30 minutes total, covering three contaminated sites with identical mechanics. Sea Wolf is the second of three Solve modules, and the full Solve battery runs 85 minutes.
Is the Sea Wolf game the same as Ocean Cleanup?
Yes. Candidates and recruiters use both names. “Sea Wolf” is the internal McKinsey name; “Ocean Cleanup” or “Ocean Treatment” shows up in some candidate descriptions. They refer to the same module.
Do I need to know biology or microbiology to pass?
No. The game uses biological-sounding labels (Aerobic, Heat Resistant, Hydrophilic, Light Sensitive) but the underlying test is pure optimization. Candidates with finance, engineering, and humanities backgrounds perform equally well when they prepare.
Can I use a calculator during the Sea Wolf game?
Not officially. All math has to be mental or on scratch paper. The averages are three-number arithmetic on a 1 to 10 scale, so a quick mental estimate is usually enough.
How does Sea Wolf compare to Red Rock and Sustainable Future Lab?
Red Rock tests analytical thinking in an ecosystem context. Sea Wolf tests constrained optimization with repeated sites. Sustainable Future Lab tests prioritization and behavioral judgment. All three measure different facets of consulting problem-solving, and all three are scored independently.
Your Next Step
The Sea Wolf game is learnable. Every mechanic is documented, every mistake is predictable, and every phase follows the same pattern across all three sites. The only question is whether you practice the four-phase flow enough times to make it automatic before test day.
Most candidates do not. They read overviews, feel informed, and walk into the Solve without ever running a timed practice site. That is why the McKinsey Solve Sea Wolf game filters out so many applicants who would otherwise be strong interview candidates.
If you want structured practice with fully playable simulations, a strategy guide, and video walkthroughs covering all four phases, the McKinsey Solve Game Guide and Simulations is built for exactly this. Candidates who complete it report an 89% pass rate based on feedback collected between January and March 2026.
You have one Solve attempt per application cycle. Use it with a plan.
About the Author
Florian Smeritschnig is a former McKinsey Senior Consultant with five years at the firm and 2,200+ mock case interviews conducted. He founded StrategyCase.com and has supported 9,000+ candidates across 70+ countries through the McKinsey Solve since November 2019. His 1-on-1 coaching clients have secured 700+ offers at McKinsey, BCG, Bain, and Tier-2 firms.
Last Updated: April 21, 2026
