A working paper for founders who can't write a $2,000,000 check — and shouldn't have to, to think at that level.

McKinsey generates roughly $16 billion a year. Not because it hires people smarter than you — its median hire is a 26-year-old with a good GPA and zero industry experience. What McKinsey actually does is take that 26-year-old and spend two years installing a set of analytical frameworks into how they think, until running them is instinct.

By the time that person sits across from a CEO, they aren't winging it. They're executing a methodology refined across fifty years and tens of thousands of engagements — and none of it is secret. Minto published the Pyramid Principle. Porter published Five Forces. Kaplan and Norton published the Balanced Scorecard. Martin and Lafley published Playing to Win. The intellectual infrastructure of elite consulting has been sitting in business school libraries for decades.

The barrier was never access to the ideas. It was the time and organizational muscle required to operationalize them on a live problem, under deadline pressure, for your specific situation — with nobody but you to do it.

That's the barrier this removes. What follows is 13 Claude skills — 12 built directly on the frameworks strategy consultancies train into their people, plus one they don't have to teach because their org chart already does it for them. Not paraphrased, not dumbed down: the intellectual foundation, the method, and a prompt built to actually run it on your business, not a generic one.

This is written for one kind of reader: the founder who is building something meant to last, doing it mostly alone, and who has better use for $2,000,000 than a slide deck. If that's you, keep going.

Strip away the brand, and a McKinsey-style engagement is three things: a team of analysts running a known methodology, on your timeline, with a partner's judgment layered on top. Two of those three things are now available to anyone with a laptop. The third — judgment, and someone's professional reputation staked on the call — is still not for sale at any price. Keep that distinction in mind for the whole rest of this piece; it's the difference between what this stack replaces and what it doesn't.

Twelve skills, four phases, one check that runs after every single one of them. This is the shape of every hard founder decision — a fundraise, a pivot, a hire, a shutdown call — whether you draw it out or not. The stack just makes each step explicit enough to actually run.

Condensed here for reading; the copy-and-paste-ready version of every prompt below ships in the attached Prompt Pack, and the installable Claude Skills ship in the attached .zip. Feed each one your real numbers, real transcripts, real documents — the frameworks are only as sharp as what you hand them.

DIAGNOSE

McKinsey Issue Tree + Minto's Situation-Complication-Question

USE WHEN Your problem statement is really a solution wearing a disguise ("we need more growth", "we need to raise", "we need a better website").

WHY THIS WORKS

Most founders skip straight to solving. Consultants are trained to distrust the first version of a problem, because it usually smuggles in an answer. A presenting problem gets decomposed into a MECE tree of 3–5 genuinely distinct hypotheses about the root cause — mutually exclusive, collectively exhaustive — and only then does the real problem statement get written, using Barbara Minto's three-part structure: the Situation (what's true now), the Complication (what changed or is threatening), and the Question (the decision this actually forces).

METHOD

FAILURE MODE TO WATCH FOR

A tree that only looks MECE. Branches quietly overlap, or one whole category of explanation (often: "the team", or "my own decisions") never appears because it's uncomfortable. Make the audit step mandatory, not optional.

McKinsey Value Driver Tree (Koller, Goedhart & Wessels) + tornado sensitivity analysis

USE WHEN You're about to build a spreadsheet before you know which 2–3 numbers actually move the outcome.

WHY THIS WORKS

The standard McKinsey valuation logic holds that a business's worth reduces to free cash flow, its growth rate, and the cost of capital discounting it — everything else (revenue, EBITDA, logos) is either a driver of those three or noise. The value driver tree maps the causal chain from an operational lever (pricing, churn, CAC, capital intensity) down to enterprise value, so that sensitivity analysis can be run on drivers instead of line items — and the model gets built around the 2–3 levers that actually matter, instead of forecasting every row with false precision.

METHOD

FAILURE MODE TO WATCH FOR

Building the full model before the tree. A beautifully precise spreadsheet that answers the wrong question is still the wrong answer — just harder to challenge because it looks rigorous.