The first stock I ever researched “seriously” took me four hours. I read the annual report, looked at the P/E ratio, checked what a few analysts said, and bought it feeling very professional.
It dropped 40% in eight months.
Not because I missed something dramatic. The company was fine. I just had no framework for understanding what I owned, what would need to happen for it to work, or what would tell me if my original reasoning was wrong. I had done research. I had not built a process. Those are different things, and confusing them is probably the most expensive mistake retail investors make consistently.
What I’m going to describe here isn’t the “best” stock selection system. I genuinely don’t believe that exists. What it is, is a system — a repeatable sequence of questions that produces consistent inputs regardless of how excited or scared I am about a particular stock on any given day. The emotional state of the investor is the variable that kills most retail returns. A system is what you put between your feelings and your order button.
Step Zero: Know What You’re Actually Trying to Do
Most people skip this entirely and it costs them everything downstream.
Before you screen a single stock, you need to answer three questions honestly. What’s your time horizon — months, years, or decades? What do you actually need from your investments — growth, income, or both? And how much of a drawdown can you genuinely sit through without panic-selling? Not theoretically. Actually.
I know a guy who told me he was a long-term investor. Sold everything in March 2020 when his account was down 28%. He’s a medium-term investor who didn’t know it yet. The system you build has to match who you actually are, not who you want to be.
These answers change which metrics matter. A ten-year holder cares about return on invested capital compounding over time. A three-year holder cares more about near-term earnings trajectory and valuation relative to growth rate. A dividend investor cares about payout sustainability above almost anything else. Build the wrong system for your actual situation and you’ll get frustrated when it keeps surfacing stocks that don’t feel right — because they don’t match what you actually want.
Write your answers down. Literally. One paragraph. Come back to it every six months to check whether you’ve drifted.
The First Filter: Business Quality Before Anything Else
I don’t start with valuation. I know that sounds backward to a lot of value investors, and I spent a few years doing it their way before I stopped. The problem with starting with valuation is that cheap stocks are often cheap for reasons that take a long time to fully understand, and by the time you understand them, you’ve either made money despite yourself or you’ve been sitting in a value trap wondering when the thesis will work.
The first question I ask about any company is whether it has any durable reason to keep making money that its competitors can’t easily replicate.
That sounds obvious. It isn’t. Most companies don’t have one. They’re good operators in competitive industries where their margins are constantly being pressured from multiple directions. Good operator plus commoditized product equals mediocre long-term returns even if everything goes “right.”
The things I’m looking for aren’t complicated to identify, just hard to find. High gross margins sustained over multiple years — not one good quarter, multiple years — because margins are confessions. A company with 65% gross margins is telling you that people pay a premium for what they make and will keep doing so. Switching costs that make customers sticky even when a cheaper alternative exists. A network that becomes more valuable as more people use it. Or simply being the low-cost producer in an industry where cost is the primary variable customers optimize for.
| Business Quality Signal | What It Looks Like | Red Flag Version |
|---|---|---|
| Gross margin | Above 40%, stable or expanding | Declining for 3+ quarters |
| Revenue retention | Existing customers spend more each year | Churn eating into growth |
| Pricing power | Raises prices without losing customers | Discounting to maintain volume |
| Return on equity | Consistently above 15% without excessive leverage | High ROE driven by debt, not earnings |
| Free cash flow conversion | FCF tracks closely with net income | Large persistent gap, FCF consistently below earnings |
None of these filters require complex financial modeling. Revenue retention you can often read from the MD&A. Gross margin is on the income statement. Free cash flow is on the cash flow statement. The analysis isn’t the hard part. The discipline to skip companies that don’t clear these thresholds — no matter how interesting the story — is the hard part.
The Second Filter: Growth That Makes Sense
After I’ve identified that a business has genuine quality characteristics, the next question is whether it’s growing and why.
Revenue growth matters. It’s the raw material that earnings eventually come from. But I’ve learned to care almost as much about why it’s growing as how fast. Growth driven by pricing power — the same number of customers paying more because the product is genuinely better or irreplaceable — compounds differently than growth driven by sales force expansion or acquisition of customers who haven’t yet proven they’ll stick around.
The specific number I anchor to varies by industry. For technology companies with high software content, I want to see 15%+ revenue growth to feel good about paying any kind of premium. For industrials or consumer staples, 7-10% is interesting. For slow-growing businesses where I’m primarily buying for income, a stable or mildly growing revenue line is fine as long as margins are expanding.
What makes me nervous regardless of sector is decelerating growth combined with an elevated multiple. Revenue growing at 30%, then 24%, then 19%, then 15% is a business on a specific trajectory, and the market generally prices that trajectory before it fully arrives. Buying a decelerating growth story at a premium multiple is a way to be technically right about the business and still lose money on the stock.
Two questions I ask every time: is growth coming from new customers or existing ones spending more? And is the growth rate changing direction? The answers to those questions tell me more than the headline number.
The Third Filter: Valuation — What You’re Paying For What You’re Getting
I use valuation third, not first. By this point I’ve already filtered for business quality and growth characteristics, which means I’m looking at a much smaller list of companies than I started with. The valuation question is different when you’ve already decided the business is worth owning — you’re asking at what price it becomes worth owning now versus waiting.
The metric I lean on most is free cash flow yield — free cash flow per share divided by the stock price. I prefer it to P/E because free cash flow is harder to manipulate than reported earnings, and it’s the number that actually tells you what the business generates for shareholders in real economic terms.
A free cash flow yield of 4-5% at current Treasury rates of around 4.4% isn’t compelling. You’re barely getting compensated for the additional risk versus a government bond. A FCF yield of 6-7% on a quality business growing cash flow at 12-15% annually is a different conversation — you’re getting a decent current yield that’s growing, attached to a business where the yield is expanding over time as earnings grow.
The PEG ratio — P/E divided by growth rate — is a rougher but faster cross-check. A PEG below 1 suggests a company might be undervalued relative to its growth rate. Above 2 suggests you’re paying a significant premium for the growth. It’s blunt, but it catches the most obvious mismatches.
| Valuation Metric | What I Look For | When I Get Interested |
|---|---|---|
| Free cash flow yield | FCF / stock price | Above 4% on quality businesses |
| PEG ratio | P/E ÷ annual growth rate | Below 1.5 for growth names |
| EV/EBITDA vs peers | Enterprise value / EBITDA | Below sector median for similar quality |
| Price-to-sales | Market cap / annual revenue | Below 5x for high-growth tech |
| Dividend yield (income stocks) | Annual dividend / stock price | Above 3.5% with sustainable payout |
The table above is a starting point, not a rulebook. Context matters enormously. A quality business in a growing market at 2x PEG might be fine. A mediocre business at 0.8x PEG might be a trap. The valuation filter tells you whether the price creates a margin of safety or requires perfection. It doesn’t tell you whether the business will succeed.
The Part Everyone Skips: What Has to Be True for This to Work
Before I buy anything, I write down — again, literally, not in my head — the two or three specific things that need to happen for the stock to work over my intended holding period.
This exercise is surprisingly hard to do honestly. It forces you to articulate your actual thesis rather than a vague sense that the company is good. “AI is a big trend and this company benefits” is not a thesis. “This company’s gross margins should expand from 58% to 65% over three years as fixed costs spread over a larger revenue base, driving free cash flow growth above 20% annually” is a thesis. One of those is checkable against quarterly results. The other is just vibes.
Writing the thesis also creates the conditions for knowing when you’re wrong. If the margin expansion doesn’t show up after two or three quarters, you have a decision to make based on something concrete rather than continuing to hold because you’re emotionally attached to your original thesis.
I also write down the one or two things that would tell me the thesis is broken — not just temporarily challenged, but fundamentally wrong. A competitor enters with a product that eliminates the pricing power. A key customer relationship terminates. Management guidance gets cut for reasons that suggest structural deterioration rather than a one-time event. Having these pre-defined means you’re responding to actual information rather than reacting to price moves.
Price moves are almost never the signal. The underlying business results are the signal.
Building the Actual Screen: My Starting Criteria
When I run a stock screen — I use Finviz and Seeking Alpha’s screener interchangeably depending on what I’m looking for — the starting criteria look like this, adjusted for the type of business I’m hunting for.
For quality growth companies:
Market cap above $2 billion — I want enough trading volume to enter and exit without significantly moving the price, and enough analyst coverage that information is reasonably priced into the stock. Gross margin above 40%. Revenue growth year-over-year above 15%. Return on equity above 15%. Free cash flow positive for at least two of the past three years. Debt-to-equity below 1.5x.
That screen across the US market in 2026 returns somewhere between 80 and 150 companies depending on how you calibrate each filter. From there, I’m reading businesses, not optimizing numbers.
For dividend and income positions:
Market cap above $5 billion — income-oriented investors want size and stability. Dividend yield above 3%. Payout ratio below 65% of free cash flow — not earnings, free cash flow. Dividend growth streak above five consecutive years. Net debt-to-EBITDA below 3x.
That screen returns a much smaller list. Maybe 40-60 companies. Many of them I’ve been watching for years and know the stories cold. The screen isn’t discovering them for me — it’s reminding me they exist and flagging when price movements have made valuations more or less attractive.
The Part Nobody Talks About: Managing the List After You Build It
A stock screen isn’t a one-time exercise. It’s something you run repeatedly and compare to itself over time.
The most useful version of a screen isn’t the snapshot — it’s the delta. What companies appeared on this screen three months ago that no longer appear? The disappearances tell you something. A company that drops off a quality screen because margins compressed or growth decelerated is a company worth understanding more carefully. Sometimes the deterioration is temporary and the stock is now cheaper for a solvable reason. Sometimes it’s the first sign of structural problems that take two years to fully manifest in the financials.
I keep a watchlist of companies that came close to making the screen but didn’t quite clear one criterion. Maybe the free cash flow yield isn’t there yet but would be if the stock pulled back 15%. Maybe the growth is right but the margins haven’t proven out. These are businesses I want to understand deeply before I need to make a decision quickly, because when the moment comes — a broad market selloff, a sector-specific overreaction to an earnings miss — I want my thinking to be a year old, not a week old.
The investors I’ve learned the most from over the years don’t spend much time on new ideas. They spend time deepening their understanding of a relatively small number of businesses until they know them better than most people in the market do. When those businesses get cheap for reasons that don’t affect the thesis, they buy. That approach is boring to describe and genuinely effective to execute.
The Mistake the System Doesn’t Protect You From
The honest thing to say after all of this is that no stock screening system protects you from the most dangerous error in investing, which is knowing a good business but not knowing a good price.
A well-constructed screen finds quality. Valuation filters narrow the price question. But between a business you’ve identified as excellent and a stock price you’ve decided is reasonable, there is still significant room for the market to prove you wrong for longer than you’re emotionally prepared for.
I’ve owned right companies at wrong prices and spent 18 months watching them go nowhere while the broad market climbed. The experience of being right about the business and wrong about the timing is its own specific kind of expensive — not because you necessarily lose money, but because the opportunity cost of that capital sitting flat is real.
The system doesn’t solve this problem. What it does is make sure that when you’re sitting in a position that’s going sideways, you can look at your original thesis and determine whether the reason you bought the stock is still intact. If it is, you wait. If it isn’t, you sell regardless of where the price is relative to your cost basis.
That last sentence is the hardest thing to do consistently. It’s also the one that separates investors who build wealth from investors who spend years hoping positions will come back to even.