I didn't set out to build a ranking model. I set out to stop opening fourteen browser tabs every time I wanted an opinion about a company.
For years my process was manual and, looking back, a little absurd. I'd pull up a ticker on Yahoo Finance and work through metrics one at a time. Free cash flow. Return on equity. Debt to equity. The PEG ratio. I'd note them down, weigh them in my head against what I knew about the business, and arrive at something like a grade. Then I'd do it again for the next company. And the next.
It worked, more or less. It was also slow, inconsistent, and — the part that nagged at me — impossible to audit. If I graded a company generously on a Tuesday and harshly on a Thursday, nothing in my process would have caught it. I was the model, and I had no version control.
Where the ideas came from
I didn't invent the underlying thinking. The metrics this site uses, and the philosophy behind how they're weighted, came from years of reading and listening to people who explain investing well. Three sources shaped it most, and each taught me something different.
The Motley Fool taught me to think about growth, and to think in years. Not what a company earned last quarter, but where the business is actually going and whether I'd want to own it long enough for that to matter. Future prospects over present optics.
Jim Cramer taught me valuation — and process. What actually makes a stock cheap or expensive, as opposed to merely low-priced or high-priced. He's also the reason I had all those tabs open in the first place: the discipline of doing the work before you buy, every time, rather than deciding first and justifying afterward.
The analysts on Seeking Alpha taught me the metrics themselves, in depth. That's where I learned what the numbers actually measure and how they mislead you — and just as importantly, an analytical style for sorting a pile of data into something you can reason about.
Here's the part I only noticed later. When I finally sat down to write the rules down properly, I realized I was building three pillars — growth, value, and quality — and I knew exactly where each of them had come from. The model didn't come out of nowhere. It's a weighted equation built from three sets of ideas I'd been absorbing for years without quite realizing they were converging.
What all of them taught me, in different ways, is the thing this whole site rests on: no single number tells the whole story. A company with a beautiful P/E might be cheap for an excellent reason. Strong revenue growth means very little if none of it converts to cash. A pristine balance sheet on a business that isn't growing is a different investment than it looks like. Every metric is a partial view, and the only way to see a company clearly is to hold several partial views at once and notice where they disagree.
That's an easy principle to state and a genuinely tedious one to practice by hand.
The tedious part
Here's what "balancing many viewpoints" actually looked like. Open the financials. Check the metric. Write it down. Open the next tab. Check the next metric. Write it down. Try to remember whether the ROE I was looking at was flattered by buybacks. Try to remember what I'd decided about a comparable company two weeks earlier, and fail, because I hadn't written that part down.
By the end I'd have a thesis and a rough grade. But the grade lived in my head and in a spreadsheet, and it wasn't reproducible. Ask me to explain why a company scored well and I could tell you a story. Ask me whether I'd applied the same standard to the company I'd looked at last month, and I honestly couldn't say.
At some point the obvious thought arrived: this is a process. Processes can be written down. Things that are written down can be run automatically.
Building it at night
The idea sat in the back of my head for a long time before I did anything with it. In July of 2025 I started sketching out plans that were actually realistic rather than fantasy. The real work began in January of 2026, when I started the genuinely hard part — converting a process I'd been running in my head into an algorithm a computer could execute the same way every single time.
I'm a pediatric physical therapist. Most of this was built after long days treating children — evenings and weekends, in the hours when the sensible thing would have been to do nothing at all.
That's not a complaint, and it's not a humblebrag. It's context for why the thing is built the way it is. When you only have an hour or two at a stretch, you don't build something clever. You build something you can put down and pick up again, that explains itself when you come back to it cold, and that doesn't depend on remembering what you were thinking last Tuesday. A lot of what I like about this model — that every score breaks into its parts, that nothing is hand-tuned, that the rules are written down rather than held in my head — comes directly from the constraint of building it tired, in short sessions, after work.
What it gives me now
The thing I wanted was never a machine that picks stocks for me. It was the end of the fourteen tabs.
Now I look at a company and get one number, built from all the metrics I'd have checked by hand anyway, weighted the way I'd have weighted them, applied identically to every company in the universe on the same day. If I want to know why the number is what it is, it breaks apart into the pillars that produced it. If I think the model is wrong about something, I can see exactly which input we disagree on.
That last part matters more than the score itself. A screen that hands you a verdict you can't interrogate is just a different person's opinion with extra steps. The point was never to outsource the judgment — it was to stop spending the judgment on data collection, so I could spend it on the part that actually needs a person.
It's the tool I wanted for myself. Publishing it came later, and it stays useful to me whether anyone else reads it or not.
The rules this model runs on, and the reasoning behind each one, are set out in full on the about page.