Lesson 3 of 7
Killing an idea in an hour
Five questions end most ideas early, each answered by one filed figure, before a model is ever built.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Five questions end most ideas inside an hour, each answered by a figure already on file, not by a feeling.
- Questions two and three are the hardest to argue around: an unsafe balance sheet, or profits that never turn into cash.
- Killing an idea fast is not a loss: it frees the hour for the next one in the log.
Most ideas do not deserve a full research pass, and an hour is usually enough to find out which ones do. Five questions, run in order, end most of them before a model is ever opened.
Five questions, and the figure that answers each
- Can I say what it does in one sentence? If not, stop (Item 1 of Form 10-K).
- Is the balance sheet safe? The current ratio and net debt against EBITDA.
- Are the returns real? Operating cash flow against net income, ten years.
- Is management aligned? Insider ownership (proxy) and insider trades (Form 4).
- Is the story already in the price? The P/E or EV to EBITDA against the sector median.
The two-minute version comes first
Lynch's two-minute drill asks for a plain-language monologue: what has to happen for the company to succeed, and the pitfalls in its path, told the way you would tell a friend. If it needs jargon or takes longer than two minutes, question one has already failed and nothing later in the hour is worth doing yet.
Where most ideas actually stop
Questions two and three are the ones a good story cannot talk its way past. A balance sheet that cannot survive one bad year rules out a story regardless of how good the story is. Returns propped up by a number that never turns into cash are not returns yet, only an estimate of them, and ten years of the comparison is sturdier than one.
Aligned enough to trust the story
Question four reads the proxy for insider ownership and Form 4 for insider trades: are the people telling the story putting their own money behind it, or selling?
Priced in is not a verdict
Question five does not say a company is cheap or expensive. It says how much of the story is already reflected in the multiple, which changes how much is left to be right about even when the story itself holds up.
Check yourself
What does Lynch's two-minute drill actually test?
- Whether the stock has been volatile over the past two minutes
- Whether you can state the reason, the plan and the risk in plain language
- Whether a broker's research note agrees with your view
Show the answer
Answer: B.
The drill is a test of understanding, not of the market. A story that needs jargon or more than two minutes has not been understood yet.Which question does a ten-year comparison of operating cash flow against net income answer?
- Can I explain what the company does?
- Are the returns real?
- Is the story already in the price?
Show the answer
Answer: B.
Cash against earnings over ten years is the check for whether reported profit is being collected, which is question three.An idea that fails question two should be
- Pushed through questions three to five anyway, since a low price might compensate
- Set aside, with the hour saved for the next idea in the log
- Researched in full anyway, in case the balance sheet improves
Show the answer
Answer: B.
The five questions exist to end ideas fast. Forcing a failed one through the rest defeats the purpose of triage.Sources
- Lynch with Rothchild, One Up on Wall Street, 1989 (millennium edition 2000), chapter 11: the two-minute drill.
What this leaves out
- The five questions triage ideas fast. They do not replace the fuller eight-step order once an idea survives them.
- The worked example is illustrative. No filed company appears in it.
- Nothing here is advice about any security.