Lesson 4 of 7
Checklists that catch mistakes
Build a short checklist from your own past mistakes and run it at the moment of decision, every time.
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
- A checklist is a short list run at the moment of decision, built from your own past mistakes, not a general strategy.
- It works by catching routine lapses under pressure, the same failure a pilot's or a surgeon's checklist catches.
- Six items or fewer, each one answerable with a fact, not a feeling.
A checklist does not tell you what to buy. It stops you from repeating a mistake you have already made once, at the exact moment you are about to make it again.
An example checklist, six questions
- Am I measuring this against the price I paid, or against the reason I bought it?
- Have I looked for the source that disagrees with me, not only the one that agrees?
- What is the base rate for a company like this achieving what I am forecasting?
- Would I buy this today, at today's price, knowing only what is filed?
- Have I written down what would prove this wrong, before deciding?
- Did I read the full risk factors section, not only the parts that support the case?
Why a short list catches what care does not
Gawande's Checklist Manifesto studied why experienced professionals, expert enough to know better, still miss routine steps under pressure and time constraints. His argument was not that they lacked knowledge. It was that knowledge does not protect against a lapse of memory or attention in the moment, and a short written list does.
Built from your own mistakes, not a generic list
A checklist copied from someone else catches someone else's mistakes. Yours should come from a short review of your own past decisions: what went wrong, and what one question, asked before you bought, would have caught it. A checklist grows one line at a time, each line earned by an actual mistake.
Short enough to run every time
A checklist with thirty items gets skipped. One with six or fewer gets run. The discipline is brevity: each question should be answerable yes or no, with something you can point to: a filing, a note, a line you wrote. Other investors' documented mistakes are a cheaper second source than waiting to make your own.
When to run it
Run it at each decision point, buying, adding or selling, at the moment money is about to move. Its job is to catch a mistake in the decision itself, not to replace the research that came before it.
Check yourself
Where should each line on an investment checklist come from?
- A guess at what could go wrong in general
- A specific past mistake of your own
- A published list of best practices
Show the answer
Answer: B.
A checklist copied from someone else catches someone else's mistakes. Each line should trace back to a mistake you actually made.Gawande's argument for checklists was mainly that experienced professionals:
- Need a checklist only when they are new to a task
- Lack the knowledge to do the task correctly
- Can still miss a routine step under pressure, regardless of how much they know
Show the answer
Answer: C.
The failure Gawande studied was a lapse of memory or attention in the moment, not a lack of knowledge. Expertise does not prevent that kind of miss.When should the checklist be run?
- Only after the position has been held for a while
- At each decision point where money is about to move
- Only at the very start of researching a company
Show the answer
Answer: B.
A checklist's job is to catch a mistake in the decision itself, so it belongs at the moment the decision is actually being made, buying, adding or selling.Sources
- Gawande, The Checklist Manifesto: How to Get Things Right, 2009.
What this leaves out
- A checklist catches routine lapses. It does not replace the analysis behind a decision or guarantee a good outcome.
- A list that grows too long stops being run. Six items or fewer is a deliberate constraint, not a suggestion to relax once it works.
- Nothing here is advice about any security.