Two things: a number for what you think the business is worth, and what would make you change your mind. Write both before you buy. The number will often be wrong. But a wrong number on paper can be checked against what happened later, and a feeling can’t, so you learn from the number.
Someone on Reddit asked what lesson to take from a trade in Meta. The stock moved after they bought, and they couldn’t decide what to do. There was no number written down, so when the price moved there was nothing to hold it up against. That’s why the decision felt impossible, and it’ll feel impossible again next time, whatever you promise yourself.
The number matters most when a company is spending big. Meta said in its second-quarter 2026 release, on July 29, 2026, that it expects capital expenditures of $130 billion to $145 billion for 2026, counting principal payments on finance leases. If you own a company doing that, write down what all that spending has to bring back for today’s price to make sense, and by when. Then check it against the annual reports two and three years out.
What would change your mind should be something you can see in a filing. For example, the cash from operations falls two years in a row and there’s no spending story behind it. Or the part of the business you bought it for starts shrinking. You pick the test when you’re calm, and the report does the judging later.
Where this breaks
It won’t tell you anything today. It keeps you from rewriting your reasons once the answer shows up.