McKinsey Sea Wolf Solver: Do Excel Solvers Work, and What to Use Instead

Is a McKinsey Sea Wolf solver worth using?

No: Excel and web "Sea Wolf solvers" exist and are sold, but using one during Solve breaks McKinsey's published assessment rules, and a tool that needs every value typed in by hand is slow on a 30-minute practice clock. McKinsey's Solve page bars applications, websites, AI tools and pre-prepared notes during the assessment, and its assessment integrity page adds a calculator to the list. The legitimate shortcut is a fast elimination method you can run in your head, which is what this page, our free PDF preview and our practice runs teach.

Try it before reading further. Road to Offer's free Sea Wolf run uses the same site, stage and trait-rule format as the full simulation. Start without an account, then create a free account to finish and keep your score, points breakdown and worked review. No payment is needed for that run. This is independent practice, not a McKinsey assessment.

For the wider test, read the McKinsey Solve guide. For the game itself and a second fully worked site, see the Sea Wolf (Ocean Treatment) guide.

What does a Sea Wolf solver actually do?

A Sea Wolf solver is a spreadsheet or web form where you type each microbe's attribute values and traits plus the site's target ranges, and it tries every three-microbe combination until one fits. That is brute force: no insight, just checking every option. We describe the category generically here; we do not name or link any product.

For a computer the search is tiny. Choosing 3 microbes from 10 candidates gives this many trios:

A spreadsheet checks 120 trios against three ranges in well under a second. The bottleneck is the input. Ten microbes with three attributes is 30 numbers, plus each microbe's traits, plus six range bounds and the trait rules: roughly 40 entries for one site. At a few seconds per entry, our estimate is two to three minutes of typing per site before the tool returns anything, and one wrong digit returns a wrong answer with full confidence.

Is using a solver in McKinsey Solve allowed?

No, using a solver during McKinsey Solve is not allowed. McKinsey's Solve page says that, except as an accommodation for a disability approved in writing by McKinsey, you may not use applications, websites, AI tools or pre-prepared notes, or record or screenshot any part of the assessment, and that you take it alone. An Excel file is an application; a web solver is a website; a printed solver output is a prepared note.

McKinsey's integrity page, which covers any assessment or interview, goes further: one screen without overlays, no calculator, and only pen and paper for notes and calculations. It also asks you to disable installed AI software that could record or assist.

McKinsey treats a breach as a violation of its values. We do not know how, or whether, a given tool is detected, and you should distrust any page that claims to know, including pages selling solvers that promise to be undetectable. The risk sits with the candidate, not the seller. The proctoring guide covers what is and is not published about supervision.

Why do solvers fail candidates even when they run?

In our analysis, solvers fail on time, typing errors, trait rules and version drift, even before the rules question. These are our own reasoning about the workflow, not McKinsey statements.

• Time. A full Sea Wolf simulation in our practice runs 3 sites, 4 stages each, in 30 minutes, which leaves about 10 minutes per site for everything. Two or three minutes of data entry per site is a large share of that, spent before any decision. McKinsey publishes no Sea Wolf-specific minutes; your invitation gives the total Solve length.
• Typos. A brute-force tool cannot tell a 7 from a 1 you mistyped. It returns a trio that fits your typed numbers, not the screen.
• Trait rules. Candidates report required and forbidden traits alongside numeric ranges. A simple sheet that only averages numbers will happily return a trio containing a forbidden microbe, which is exactly the trap in the worked example below.
• Version changes. McKinsey does not publish a Sea Wolf rulebook, and a tool built on one reported layout breaks when the screen shows a different structure. A method adapts; a template does not.

What is the fast manual method instead?

The fast method is to eliminate before you calculate: drop microbes with a forbidden trait, lock in the required trait, drop any microbe that makes a range impossible, then turn each average range into a sum window and check the few pairs left. Here is an original fictional site, not McKinsey content.

Fictional site: Kelp Point. Average capture must be 6 to 7, average endurance 4 to 5, average spread 3 to 4, all inclusive, for a trio of three microbes. The trio needs at least one salt-tolerant microbe and no toxic microbe.

Seven microbes give C(7, 3) = 35 possible trios. Watch that number fall.

Step 3 in numbers. An average range of 6 to 7 over three microbes means the trio's capture must total 18 to 21. A already supplies 8, so the other two must total 10 to 13. The same move for the other attributes:

C has spread 8, so its partner would need spread 0 or less to stay within 5 to 8. Every remaining microbe has spread of at least 3, so C is out. That leaves D, E, F and G: six pairs, each needing three additions you can do at a glance.

Only D + F fits, so the treatment is A, D and F. Confirm it once with the averages:

Notice A on its own fails two ranges: capture 8 is above 7 and endurance 2 is below 4. Screening single microbes against the ranges would have thrown out the only salt-tolerant option and left no valid trio at all. Averages are about the group.

There is a second trap: D, E and G average 7, 4.67 and 3.67, inside every range, yet none of them is salt-tolerant. A numbers-only spreadsheet would offer it; the trait check rejects it in one look. Our full brute-force check of all 35 trios confirms A, D and F is the only valid treatment at this fictional site.

Where can you get the method as a PDF?

You can get it from Road to Offer's step-by-step Sea Wolf PDF guide: a free account unlocks a 6-page preview at the free Solve guides page, and the full PDF comes with the Sea Wolf pack, the Solve pack or qualifying Annual access. The PDF is the legitimate version of what a solver promises: a repeatable order of checks you practise until it is fast, then carry into the test in your head, with pen and paper.

Then test the method under time. The free Sea Wolf run ends with a worked review that shows which constraint each decision met or missed. Paid Learning mode, with coaching as you go and untimed play, comes with the Sea Wolf pack, the Solve pack or qualifying Annual access. If you have an invitation date, a free account also lets you build a Solve plan from your invitation length and test date, up to 28 days.

Plan your Sea Wolf practice up to test day

Choose your invitation length and test date for dated practice steps selected for that invitation, up to 28 preparation days. A free account is required to view and save the plan.

Build my Solve plan

What about Red Rock and Ecosystem solvers?

The same answer applies to Red Rock and Ecosystem solvers: outside tools are barred during Solve, and neither game rewards one. Candidates describe an on-screen calculator inside Red Rock and data that changes as you move through the task, so a pre-built spreadsheet has little to work on; the Red Rock study guide covers the reported interface and a method for its sums. Ecosystem Building does not appear in 2026 candidate reports, so an Ecosystem solver targets a game you are unlikely to see; the Ecosystem guide explains what it was and where its skill is tested now.

Sources and evidence boundaries

• Official (checked November 3, 2026): McKinsey's Solve page for the ban on applications, websites, AI tools, prepared notes, recording and outside help, the written accommodation exception and the consequences; McKinsey's assessment integrity expectations for one screen, no calculator, pen and paper only, and disabling AI software.
• Reported by candidates: Sea Wolf site ranges, three-microbe averages, and required and forbidden traits; Red Rock's on-screen calculator; Ecosystem Building's absence from 2026 reports.
• Our analysis and data: the Kelp Point site, its numbers and the brute-force check are original fictional teaching material. Data-entry time estimates are our reasoning, not measurements. The 3 sites, 4 stages and 30-minute clock describe Road to Offer practice, not McKinsey's test.
• Not published: how McKinsey detects outside tools, Sea Wolf's minutes, score formula or pass mark.