Last Updated on March 27, 2026
The McKinsey Problem Solving Game (PSG), now universally referred to as McKinsey Solve, has been part of the Firm’s recruiting process since late 2019. It assesses analytical reasoning, structured problem solving, and decision-making under time pressure through interactive simulations rather than traditional test questions.
Since StrategyCase launched the first global Solve preparation resources in 2019, we have continuously tracked game updates through direct candidate reports, controlled test replications, and simulation benchmarking. While McKinsey and its game-development partner have tested multiple experimental modules over the years, only a subset ever reaches consistent production deployment.
One of the more recent additions to the Solve environment is the Sea Wolf game, internally also referred to as Ocean Cleanup or Ocean Treatment. Unlike early test-phase descriptions circulating in 2023 and early 2024, Sea Wolf is now a fully standardized production game with stable mechanics and a consistent placement inside the Solve sequence.
This article reflects the current live version of Sea Wolf as encountered by candidates today.
Where Sea Wolf Sits in the Solve Sequence
In the current Solve configuration, candidates typically encounter two to three games in one sitting:
- Red Rock Simulation
- Sea Wolf (Ocean Cleanup / Ocean Treatment)
- Sustainable Future Lab (new and not fully rolled out)
Sea Wolf appears as the final module. There is no beta disclaimer, no experimental labeling, and no variability in whether candidates receive it. It is now part of the standard Solve battery in regions where the two-game sequence is deployed.
Total test time remains 65 minutes across all modules (85 minutes if the new game is part of the assessment). Sea Wolf itself runs 30 minutes.
Overview of McKinsey Sea Wolf Game
The game simulates environmental remediation through microbial engineering and challenges candidates to design optimal treatments for plastic-contaminated ocean sites.
The Sea Wolf module lasts 30 minutes and consists of three contaminated ocean sites.
Each site has a different environmental profile, but the underlying mechanics remain identical. In practice, candidates solve the same core optimization problem three times with varying inputs. This repetition is deliberate and tests whether candidates learn and improve their approach as they progress, similar to the multi-level structure used in older versions of Plant Defense.
Objective of the Sea Wolf Game
The goal in Sea Wolf is to design an effective microbial treatment for each contaminated site.
For every site, candidates must identify three microbes whose averaged attributes and collective traits best match the site’s required treatment profile. The closer the match, the higher the treatment effectiveness score.
Success depends not on trial-and-error, but on systematically translating site requirements into structured filtering and selection logic.
High-Level Game Flow
For each of the three sites, candidates follow the same structured process:
- Filter relevant characteristics
- Categorize microbes based on attributes and traits
- Narrow down to promising candidates
- Submit a final three-microbe treatment combination
The core challenge is aligning microbe properties with site requirements while managing time across all three sites within the 30-minute window.
A short tutorial precedes the first site to introduce the interface and mechanics, similar in function to the Red Rock tutorial.
Interface and Phase 1: Filtering Characteristics
When the Sea Wolf scenario begins, candidates see an overhead ocean map with three contaminated sites. Only the first site is active initially.
At this stage, no microbes are selected yet. The task is to interpret site requirements and configure filters.
Site requirements panel (top right)
This is the most important reference panel on the screen. It displays the treatment requirements for the active site, including:
- Three numerical attributes with required ranges
Example:
Density: 6–8
Energy: 2–4
Size: 8–10 - Two traits
One marked as desired
One marked as undesired
These requirements remain visible throughout the site and should anchor every decision.
Characteristics selection panel (left)
On the left side of the screen is the Characteristics panel. This is where candidates configure which characteristics to prioritize for filtering.
It contains:
Attributes
- Three attribute names (for example Density, Energy, Size)
- Each with an activation toggle
- A numerical range slider from 1 to 10
Traits
- Four possible traits (for example Aerobic, Heat Resistant, Hydrophilic, Light Sensitive)
- Each with an on/off toggle
Candidates must select exactly two characteristics in total, across attributes and traits. Any combination is allowed: two attributes, two traits, or one of each.
At this stage, this panel does not select microbes. A prompt in the interface reinforces this: “Select 2 microbe characteristics according to your current site information.”
The test here is whether candidates can translate site requirements into relevant filtering criteria, rather than jumping prematurely to solution building.
We have adjusted our guide to include strategies for the Sea Wolf game and also created practice games for you that replicate the actual game accurately. For more information on our Solve Game Guide and Simulations, see below.
Phase 2: Microbe Evaluation and Shortlisting
Once characteristics are configured, the microbial pool becomes visible. Each microbe is described by:
- Numerical values for the site’s attributes
- Binary presence or absence of traits
Based on our strategy candidates should now:
- Eliminate microbes violating site attribute ranges
- Exclude microbes containing undesired traits
- Ensure at least one candidate microbe possesses the desired trait
Through this process, candidates narrow the pool to a small set of promising microbes that can realistically form an effective treatment. Unsuitable microbes can either be sent to Site 2 (based on the now available information for Site 2) or returned if it does not fit either site.
This is where structured filtering and disciplined elimination matter. Poor filtering in this phase inevitably leads to weak final treatments.
Phase 3: Build the Prospect Pool
After configuring filters and completing the initial distribution, you move into an in-depth optimization phase for Site 1.
At this point, the game introduces a controlled selection process to construct your prospect pool, which will later serve as the basis for the final treatment decision.
How the prospect pool is built
- You start with six microbes already in your pool.
- You are then presented with three candidate microbes at the top of the screen.
- You must select exactly one of them.
- This process is repeated four times, expanding the pool from six to ten microbes total.
The interface is structured as follows:
- Top: three candidate microbes for the current round
- Bottom: your growing prospect pool
- Top right: the site characteristics panel, fixed in place, showing target attribute ranges and desired/undesired traits
This panel remains your primary reference point throughout the phase.
How to approach each selection
Each choice should follow a consistent, structured logic:
Evaluate attribute impact
Consider how the microbe’s attributes will influence the future averages of the final three-microbe solution, not just whether the values look reasonable in isolation.
Exclude undesired traits
If possible, avoid microbes that carry traits explicitly marked as undesired for the site.
Prioritize the desired trait
If one candidate includes the site’s desired trait, this is often a strong signal in its favor, especially if your pool does not already contain it.
Critical mindset shift
None of the candidates will consistently be “perfect.”
You are forced to choose one option in every round, even when all three have drawbacks. This is intentional. The game is designed to test optimization under constraints, not idealized selection.
Key implications:
- Do not optimize individual picks in isolation.
- Optimize the portfolio, not the single microbe.
- A 100% outcome is not always mathematically achievable.
- Accepting an imperfect microbe can be correct if it stabilizes the overall attribute balance.
- If your running averages are trending too low, selecting a microbe above the target range may be the optimal corrective action.
Your objective is not to assemble a collection of “good” microbes, but to enable the best possible final combination of three.
Effectiveness ceiling
The overall treatment effectiveness score is capped.
Depending on the site’s constraints and the available microbes, the maximum achievable score typically falls between 80% and 100%. In some scenarios, no perfectly matching solution exists.
Many candidates underperform by continuing to search for a theoretical perfect configuration long after a strong solution is already available.
Knowing when to stop optimizing and move on is part of what the game evaluates.
Phase 4: Finalize the Treatment
Once your prospect pool of ten microbes is complete, you enter the final decision stage.
Your task is to select three microbes whose:
- averaged attributes fall within the site’s required ranges, and
- collective traits include the desired trait and avoid undesired ones where possible.
This selection becomes the final contamination treatment for the site.
Progression through the game
After confirming your treatment:
- You proceed to Site 2, repeating the same four-phase process.
- Then to Site 3, which completes the module.
The mechanics do not change. Only the site profiles differ.
Feedback in StrategyCase simulations
In StrategyCase simulations, the game concludes with a comprehensive feedback report that goes far beyond a numeric score.
The report provides:
- A breakdown of your decision logic
- How your selections evolved across rounds
- Where your filtering and optimization were effective
- Where structural weaknesses occurred
- How closely your actions aligned with the site objectives
Most importantly, the feedback is actionable. It translates performance signals into concrete improvement areas, allowing you to refine your preparation efficiently rather than guessing what went wrong.
What the Sea Wolf Game Tests
Sea Wolf assesses three core abilities:
- Translating environmental constraints into structured selection criteria
- Filtering feasible options under multiple simultaneous constraints
- Performing quantitative optimization under time pressure
Unlike Ecosystem (systems interdependencies) or Red Rock (resource allocation and dynamic optimization), Sea Wolf is a multi-constraint selection and averaging problem disguised as a biological cleanup scenario.
The same logic is applied to all three contaminated sites. Each site presents new attribute ranges and trait requirements, but identical mechanics.
This repetition tests:
- Whether candidates learn from earlier sites
- Whether their filtering and selection become faster
- Whether they maintain time discipline across the full 30 minutes
This mirrors the learning-curve evaluation logic previously used in older multi-level Solve modules.
Time Management Guidance
Although candidates may spend as much time as they wish on each site, total module time is fixed at 30 minutes.
A practical allocation is:
- Site 1: ~12 minutes (learning curve) – if you practice with our simulations, the learning curve happens before that. Our candidates report that they need around 7 minutes per site.
- Site 2: ~9 minutes
- Site 3: ~9 minutes
Failing to move on in time is one of the most common causes of underperformance.
Ace the McKinsey Sea Wolf Game
- Crack every game: Proprietary guide and video insights detailing the exact steps and strategies used by successful candidates
- Score high: Tailored tactics and gameplay walkthroughs based on real test-taker feedback
- Prepare efficiently: Focus on what matters most and avoid wasted preparation time with proven methods to master all required skills
- Interview-ready bonus: Includes a free 14-page McKinsey Interview Primer with essential guidance for case and PEI preparation
*Based on customer feedback from November to December 2025
Latest update: January 2026
Our Credentials
- 9,000+ candidates supported from more than 70 countries since November 2019
- 600+ test-taker interviews informing continuous refinement, combined with expert game designer input and firsthand McKinsey experience
- Complete preparation suite including fully playable game simulations, a 129-page strategy guide, automated Excel tools for Ecosystem Creation, and a video course covering gameplay and winning strategies
- 100% proprietary content
Click on the image below to learn more about our Solve Game Guide and Simulation.
McKinsey Solve Game Guide 23rd Edition
SALE: $169 / $99
