The Playbook · Framework

How to Think About “What Would Buffett Buy?”

The useful version of the question is not which ticker Buffett would pick — it's the four filters he would run any business through.

7 min read · Published 2026-07-16

Search any stock's name plus “Buffett” and you'll find articles claiming he would love it. Most of them work backwards: pick a popular stock, then decorate it with Buffett quotes. That's entertainment, not analysis.

The honest version of the question is a process, not a pick. In his 1977 shareholder letter, Buffett laid out the four things he wants in any business, and they haven't changed since: a business he understands, with favorable long-term economics, run by able and honest people, at a sensible price. Everything else in his letters is elaboration on those four filters.

Fig. 1 — Buffett's four filters

Every idea must clear all four, in order. Most fail at the first filter — and that is the system working, not failing.

Filter 1: Can you understand it?

Understanding doesn't mean you can use the product. It means you can describe, with numbers, how the company turns revenue into cash today and why that will still be true in ten years. Who pays, how often, what stops them from leaving, and what the business must spend to keep its position.

If a business fails this filter for you, it fails — even if it would pass for someone else. Buffett skipped most technology stocks for decades not because they were bad businesses but because he couldn't predict their tenth year. That discipline cost him some winners and spared him many losers.

Filter 2: Are the long-term economics favorable?

Favorable economics show up as high returns on the capital the business actually employs, sustained over time, with a credible reason — a moat — for why competitors haven't been able to compete those returns away.

Beware of good results with no moat behind them. A commodity business at the top of its cycle can look brilliant for two years. The test is not the best year but the worst plausible one: does the company still earn acceptable returns when the environment turns hostile?

Filter 3: Is management able and honest?

Read three years of shareholder letters back to back. Do the promises made in year one get honestly accounted for in year three? Managers who talk in owner's terms — per-share value, returns on retained earnings, honest mistakes — tend to run companies the same way.

Then check capital allocation. Track where the cash went: reinvestment, acquisitions, buybacks, dividends. Buffett's rule of thumb is that retaining a dollar of earnings is only justified if it creates at least a dollar of market value. Many managers fail exactly this test.

Filter 4: Is the price sensible?

Only after a business clears the first three filters does price enter the picture. Buffett's shorthand: it is far better to buy a wonderful company at a fair price than a fair company at a wonderful price. Fair is not the same as any — the price still has to leave a margin of safety against your own errors.

Note the order. Starting with price — screening for whatever looks statistically cheap — tends to fill a portfolio with businesses that are cheap for a reason. Buffett spent his early career doing exactly that with Graham's 'cigar butt' method and moved past it under Charlie Munger's influence.

Takeaway: Stop asking which stock Buffett would buy. Ask instead: do I understand it, is it protected, is it honestly run, and does the price forgive my mistakes? Any business that clears all four filters is a candidate — whoever's portfolio it is or isn't in.