The problem screening solves
There are thousands of listed companies. You cannot research them all, and picking the ones you happened to hear about means your plan is built from whatever marketing reached you.
A screen is a set of filters applied to facts every public company must publish. Ask for companies that made a profit in each of the last four quarters and thousands become hundreds. Add a debt limit and hundreds become a few hundred. Add a size floor and you have a list you can work with.
What you end up with is not a list of good investments. It is a list of companies that are not obviously disqualified, which is a genuinely useful thing to have and a much more modest claim.
The facts worth filtering on
Profitability asks whether the company makes more than it spends. A company with years of losses may still be a fine investment, but it is a bet on a future rather than a claim about a present, and beginners are usually better served by companies that already work.
Debt asks what happens when conditions get difficult. A company whose earnings comfortably cover its interest payments has room to survive a bad year. One that is barely covering them does not.
Size asks how much a single piece of bad news can move the price. Larger companies move less violently and are followed by more people, which means fewer nasty surprises even though it also means fewer spectacular gains.
Valuation asks what you are paying for what you get. A wonderful company at an absurd price is a poor investment; a mediocre one cheaply bought can be a good one.
A worked screen
Start with roughly 5,000 listed companies. Ask for those that made a profit in each of the last four quarters, and you might be at 1,800. The screen has already removed the companies that are stories rather than businesses.
Add a requirement that earnings comfortably cover interest payments, and perhaps 640 remain. Add a size floor of $10 billion, and you might reach 87.
Eighty-seven is a workable number. From there a plan might hold the eight that rank best on whatever you care about most. Notice that every step was a fact from a filing rather than a forecast or an opinion, which is exactly what makes the result testable.
How screens mislead
Filters use published numbers, and those numbers describe a quarter that has already ended. A company that looked healthy on paper may have deteriorated since, and no screen can see that.
Numbers also mean different things in different industries. A software company and a utility have completely different normal levels of debt, so a single debt filter applied across everything will systematically exclude entire sectors for reasons that have nothing to do with quality.
And a screen with too many filters produces a very short list of companies that happen to satisfy every condition at once. That is not rigour, it is a small sample dressed up as selectivity, and it usually means the plan has been fitted to the present moment.
What a screen is not
It is not a recommendation. It is a filter, and eighty-seven companies passing your conditions says only that they were not excluded by them.
The list is the beginning of the work. Those companies still need rules for when to buy them, when to sell them, and how much of the account each may take, and that whole arrangement still needs testing across years.
A screen without a tested plan behind it is a shopping list with no budget and no exit. That is how people end up holding twenty companies they cannot explain.
Screening in Stax
You can build a screen by describing what you want in ordinary language rather than knowing which metric names to type, and the count updates as each condition is added so you can see the effect of every filter.
The important part is what happens next: the shortlist hands straight to the plan builder and then to the history test, so a screen never dead-ends as a list you admire and do nothing with.