Most candidates "practice Sea Wolf" by memorizing attribute labels and hoping the live sitting looks familiar. That is the wrong target. Sea Wolf sits inside Solve, McKinsey's gamified assessment, and the failure mode I see is not weak biology. It is averaging away a hard reject because three other attributes look strong.
The thesis is simple: a useful Sea Wolf simulator must force filter-then-optimize under a clock, then let you score filter errors separately from optimization errors. Rules, site pacing, and reported scoring folklore belong in the sibling Sea Wolf guide. This piece is about how to practice with a simulator, what "good" looks like, and how to use CoachNed's timed Kestrel Shelf drill at /mckinsey-sea-wolf without pretending it is McKinsey's item bank.
After reading, you should be able to audit any practice tool in two minutes, score your own runs with a four-line card, and run a five-day plan that transfers to the live assessment.
What Solve actually publishes about Sea Wolf practice
McKinsey's public line on Solve is consistent and deliberately thin. On the official Solve page, the firm describes a gamified assessment of problem-solving ability, says there is no need to prepare, and states that performance is considered with the rest of the application. The interviewing overview places Solve as a step for most consulting roles and repeats the same framing.
Duration is not a single public constant. McKinsey's Solve FAQ (as of the August 2025 PDF) says the length is outlined in your invitation email, each task starts with an untimed tutorial, and the assessment runs on a PC or Mac after a tech check. Your invite is the rulebook.
In a 2024 interview with the game-based innovation lab, McKinsey names Sea Wolf and describes selecting microbes to clean contaminated sites as a way to evaluate analytical thinking in context. That confirms the premise. It is not a scoring matrix, pass mark, or fixed module list.
Integrity is not folklore. McKinsey requires you to participate alone and forbids applications, websites, AI tools, pre-prepared notes, recordings, and screenshots except as an approved disability accommodation. Practice before the window; during it, follow the tutorial only.
What McKinsey does not publish is a Sea Wolf scoring formula. Candidate reports about three sites, thirty minutes, and efficiency deductions live on forums. Useful for orientation. Not primary fact. If a simulator claims "exact McKinsey scoring," treat that as marketing.
The four moves a simulator must force
When I debrief candidates who froze on Solve, the same gap appears: they practiced optimization without practicing rejection. A Sea Wolf-style exercise is a multi-constraint selection loop. A simulator earns its keep only if it reproduces pressure on all four moves below.
| Move | What you do | What flimsy practice skips |
|---|---|---|
| Rule capture | Rewrite site conditions as a checklist before touching options | Reading the briefing once and "remembering" it |
| Hard reject | Eliminate any option that fails a mandatory rule | Keeping near misses because other attributes look strong |
| Survivor compare | Rank remaining options on the same order every time | Switching which attribute you stare at mid-pass |
| Set validation | Recheck the final combination as a set, not as three solo wins | Submitting three individually pretty picks that break together |
If a tool only quizzes "which microbe has the highest salinity tolerance," it is a sorting drill wearing a Sea Wolf hat. The hard-reject step is what most free tools underweight: a deal that fails a hard constraint is dead even if the IRR looks lovely.
Ned's rule. Write the disqualifiers first, in your own words, before you touch a candidate. If you optimize first and filter second, you will keep a near miss because it looks expensive to abandon. Interviewers and scoring engines both punish that habit.
How CoachNed's sampler maps to those four moves
CoachNed's Sea Wolf practice at /mckinsey-sea-wolf is an original, independent micro-study. It is not McKinsey's game, not a clone of confidential Solve content, and not a promise of any score. It is a short stand-in for the filter-then-optimize feel: eight named microbes, two hard reject rules, a Site 1 pool, a three-microbe treatment choice, and a seven-minute clock.
The setting is the Kestrel Shelf spill. Each microbe shows three numeric attributes (P, M, E) plus a toxin flag. Energy is the sum of the three attributes. The briefing is compact on purpose: reject toxins; reject energy of 5 or less; keep the rest on Site 1; from Site 1, pick the three highest-energy non-toxin microbes as the treatment.
No account is required to start. Everything on CoachNed is open for seven days with no card; then $120 for a recruiting season or $49 a month.
Honest gaps matter. The sampler is one site, not a reported three-site Solve sitting. It uses a transparent energy rule rather than McKinsey's unpublished attribute averages and trait filters. It scores reject accuracy and treatment overlap, not McKinsey's process model. Use it to rehearse checklist writing, hard rejects, consistent comparison, and a deliberate set check. Use the sibling guide for format lore. Use the Red Rock and SFL simulators for the other Solve skill families.
Worked example: score a Kestrel Shelf run like an interviewer
Reproduce this on paper in under three minutes, then compare to the sampler's walkthrough.
| Microbe | P | M | E | Energy | Toxin | Correct bin |
|---|---|---|---|---|---|---|
| Aurelia | 2 | 3 | 1 | 6 | no | Site 1 |
| Kestrel | 3 | 2 | 2 | 7 | no | Site 1 |
| Nimbus | 1 | 1 | 3 | 5 | yes | Reject |
| Orion | 3 | 1 | 1 | 5 | no | Reject |
| Pebble | 2 | 3 | 2 | 7 | no | Site 1 |
| Quill | 1 | 2 | 2 | 5 | no | Reject |
| Ridge | 3 | 3 | 1 | 7 | no | Site 1 |
| Sable | 2 | 1 | 3 | 6 | yes | Reject |
Reject set: Nimbus and Sable (toxin), Orion and Quill (energy 5). Site 1: Aurelia, Kestrel, Pebble, Ridge. Treatment: Kestrel, Pebble, and Ridge at energy 7. Aurelia at 6 is the correct leftover.
A candidate who submits that set has the arithmetic. I still ask: did they write both reject rules before placing anyone? Did they treat energy 5 as a hard fail, or keep Orion because P equals 3 looks strong? Did they build treatment only from Site 1 survivors?
| Answer pattern | Likely break | Fix on the next rep |
|---|---|---|
| Keeps Orion on Site 1 | Softened a hard energy floor | Circle every mandatory threshold before sorting |
| Rejects Aurelia | Confused "low on one attribute" with "fails energy" | Compute the sum; do not invent a third rule |
| Treatment includes Sable | Optimized energy before checking toxin | Filter toxins off the board permanently |
| Treatment is Aurelia, Kestrel, Pebble | Ranked by one attribute or name order | Sort by the stated objective, then recheck the set |
Say the objective aloud before you submit: "highest energy, no toxins, already on Site 1."
How to score yourself without a published pass mark
McKinsey will not give you a Sea Wolf threshold. Do not invent one. After each timed run, fill this four-line card before you look at any walkthrough.
| Line | Grade 0 or 1 | Pass look |
|---|---|---|
| Checklist | Wrote every hard rule in your own words first | Two or more explicit reject statements on paper |
| Hard rejects | No near miss kept for soft reasons | Zero mandatory failures in the keep pile |
| Compare order | Used the same attribute order for every survivor | No mid-pass switching of which column you trust |
| Set check | Rechecked the final three as a combination | One spoken sentence stating the objective before submit |
Four out of four means the method held even if one click was wrong. Two out of four with a lucky treatment means you got a score you cannot reproduce. Separate error types when you miss: reading (rewrote a range), comparison (mis-ranked survivors), or combination (three valid picks that fail together). Fix only one type on the next rep.
What transfers to the real sitting, and what does not
| Transfers | Does not transfer |
|---|---|
| Writing constraints as a checklist under time | Exact labels, art, or microbe names from any practice tool |
| Treating mandatory rules as binary | Any third-party "efficiency percent" as McKinsey's score |
| Consistent compare order across options | Assuming your invite has the same module mix as a forum post |
| Rechecking a final set before submit | Memorized answers; Solve varies parameters across candidates |
| Calm pacing after a first wrong click | Using notes, AI, or a second screen during the live window |
McKinsey's careers blog on Solve's design describes permutation-heavy scenarios so candidates do not share identical experiences. A memorized board does not transfer; a portable filter-then-optimize routine does.
If your invitation also includes Red Rock or SFL, switch skills deliberately. Sea Wolf rewards multi-constraint selection; Red Rock rewards exhibit math; SFL rewards prioritization. The MBB assessment simulator roundup maps the free options.
A five-day practice plan that fits a busy week
| Day | Assignment | Success look |
|---|---|---|
| 1 | One untimed Kestrel Shelf run; write the two reject rules before any click | Checklist exists before Site 1 has a name on it |
| 2 | One timed run at /mckinsey-sea-wolf; fill the four-line card before the walkthrough | Card completed in under two minutes |
| 3 | Replay only the miss type from Day 2; change one behavior | Same checklist order, one new checking action |
| 4 | Timed run, then ten minutes of /drills/structure turning prose into branches | Structure drill feels like the same reject habit |
| 5 | Full dress: tech-check your laptop, close other tabs, one clean timed run, stop | No mid-run rule rewriting |
If Day 5 still shows soft rejects, fix the checklist rather than adding tools. If it is clean but slow, shorten the compare pass, not the rule-writing pass. For the same habit in spoken cases, use /start or /interview.
Practice this today
Open /mckinsey-sea-wolf. Before you place a single microbe, write on paper: "Reject toxin. Reject energy 5 or less. Treatment equals top three energy on Site 1." Run the seven-minute clock. When it ends, fill the four-line card before you read the walkthrough. Then open the Sea Wolf guide only for format context you still lack, not as a substitute for another timed rep.
CoachNed is independent and is not affiliated with McKinsey & Company.
Frequently asked questions
Is there a free McKinsey Sea Wolf simulator?
CoachNed's Kestrel Shelf drill at /mckinsey-sea-wolf is free to start with no account. It is original practice for the filter-then-optimize skill, not McKinsey's assessment and not a copy of confidential Solve content.
How should I practice for Sea Wolf without memorizing answers?
Write the hard rules first, run under a clock, and score reject errors separately from treatment errors. McKinsey designs Solve so parameters vary across candidates, so memorized boards do not transfer.
Do I need biology knowledge for Sea Wolf?
No. McKinsey states that Solve does not require previous business knowledge or gaming experience, and its own description of Sea Wolf frames microbe selection as a way to evaluate analytical thinking, not domain expertise.
How long should I spend practicing with a Sea Wolf simulator?
Five timed runs with a written scorecard beat twenty untimed clicks. If you can produce four out of four on the card above and finish cleanly once under a clock, spend remaining hours on cases and the other Solve modules your invite lists.
Will a high simulator score get me through Solve?
No third-party score predicts McKinsey's unpublished model. Recruiters consider Solve with the rest of your application. Use simulators for method rehearsal, then follow the tutorial and integrity rules in your live sitting.
Sources
- Solve, McKinsey's assessment game — official description, preparation stance, integrity rules, and how Solve sits with the rest of the application. Checked 2026-09-24.
- Interviewing at McKinsey — places Solve in the consulting interview process and restates the no-prep framing. Checked 2026-09-24.
- From apprenticeship in space to selecting microbes — McKinsey names Sea Wolf and describes microbe selection for contaminated sites as an analytical-thinking task. Checked 2026-09-24.
- Innovating recruiting through online gaming — historical design note on permutation-heavy scenarios so candidates do not share identical experiences. Checked 2026-09-24.
- Solve FAQ (August 2025 PDF) — invitation-email duration, untimed tutorials, PC/Mac tech check, and variation across scenarios. Checked 2026-09-24.