How the analysis becomes a decision

Post #1 laid out the ladder this whole thing climbs: raw data at the bottom, a decision at the top. This is the same idea made concrete — the actual machinery that turns a pile of analysis into a single call on a stock.

Every company I track gets the same treatment, because I'm not collecting data for its own sake. I'm trying to get to one question — buy, sell, add, or trim — and everything below exists to answer it.

The seven elements

It starts with a full report on the business, broken into seven elements:

• Segments — what the business actually is, and where the money comes from.
• Moat — whether it can defend what it earns.
• Guidance — what management said to expect.
• Key variables — the handful of things that actually drive the outcome.
• Growth — where it's headed, and how fast.
• Valuation — what you're paying for it.
• Forensics — the accounting and governance check: is the picture real?

Seven reads on one company. On their own they're just description. The work is turning them into a decision.

First, a pre-check

Before any scoring, four of those elements act as a gate: forensics, segments, key variables, and guidance. The logic is simple — there's no point scoring a business you don't trust or don't understand. If the forensics throw a flag, or the segments don't add up, or the key variables aren't clear, everything downstream is built on sand. So these come first: a trust-and-understanding check before the machine starts assigning numbers.

Then, the scores

What clears the gate gets scored, on four components:

• Industry strength — the quality of the pond it swims in.
• Quality (the moat) — how good the business itself is.
• Growth — the trajectory.
• Valuation — the price.

These roll into one composite score — not a black box you take on faith, but a number you can open up to see which component is carrying it and which is dragging.

From score to decision

A score still isn't a decision. Two things turn it into one:

• Direction. Not just where the score sits, but where it's heading — the trajectory I wrote about last time, and the swing variables that bend it. A 7 on the way up is a different stock from a 7 on the way down.
• Reward asymmetry. What you make if it works versus what you lose if it doesn't. A good business at a bad price is still a bad decision.

Score, plus direction, plus reward asymmetry → the call. Buy, sell, add, trim.

That's the whole stack, bottom to top: seven elements → a trust-and-understanding gate → four component scores → one composite → and finally, with direction and asymmetry, a decision. Every layer is meant to be visible, so you can disagree with the call and see exactly where you'd disagree.

None of this is fully built yet — the lower layers are further along than the top. But this is the blueprint. The point was never to hand you a verdict; it's to show you the whole stack that produced one.