Most people who say they have an investment thesis have a preference - a list of things they like the sound of. The difference is whether anything could show it to be mistaken.
10 min read · Level 2
I have come to think the falsifier is the entire thesis and the rest is preamble, which is close to the opposite of how theses are usually taught and written. The belief and the reasoning are the parts people enjoy - they are where the intelligence shows, and they are what gets presented. But a belief with no falsifier attached is not a position an investor can be held to, and in practice it does not survive contact with an attractive opportunity: something will be found that fits, because with no stated boundary everything fits eventually. I would go further and say that if you are going to write only one line, write the falsifier and skip the belief, because the falsifier implies the belief and cannot be faked in the same way. The reason this is uncomfortable is that writing a falsifier is the moment you commit to being able to lose an argument with yourself later, and most of us would rather keep the option open. That is exactly why it is the valuable part.
Ask ten people what their investment thesis is and most will answer with a list: "I like founder-led companies in emerging markets, particularly in fintech, with strong unit economics." That is a preference. It describes a taste, and a taste cannot be wrong - it can only be unfashionable.
A thesis is a claim about the world that could turn out to be false, stated clearly enough that you would know if it had. It has three parts:
The belief what you think is true about a market or a change The reasoning why you think it, and what evidence supports it The falsifier what you would have to observe to conclude you were wrong
The third part is the one almost nobody writes down, and it is the part that does all the work. Without it, every outcome confirms you. A position that goes up proves the thesis; a position that goes down proves the market has not understood it yet. Both readings are available at all times, which means the belief is not doing any work at all.
Compare two statements about the same sector:
Preference "I like vertical software for small businesses." Thesis "Small-business software wins on distribution, not features, because the buyer has no procurement process and no time to evaluate. So the companies that grow are the ones embedded in an existing workflow the owner already uses daily. I would be wrong if the fastest-growing companies in this sector over the next three years turn out to be ones that won on product depth against an incumbent."
The second is longer, and length is not the point. The point is that in three years there is a fact of the matter, and it is knowable. You can be held to the second statement. You cannot be held to the first.
A thesis is also narrower than most people find comfortable. "I invest in technology" is not a thesis because technology is not a claim. The narrowing is what creates the value: it is what lets you know something that the average buyer of that asset does not, which is the only durable reason to expect a better-than-average result. If your belief is the consensus belief, it is already in the price, and you are not being paid for holding it.
That is where the two ideas connect. Your circle of competence tells you where you are capable of forming a view; the thesis is the view itself. A thesis outside your circle of competence is a guess with a paragraph attached.
Because the main job of a thesis is not to help you say yes. It is to let you say no quickly, cheaply and without guilt.
An investor without a thesis evaluates every opportunity from scratch, on its merits, in isolation. This sounds open-minded and is exhausting. It means each pitch gets the same hours, the polished ones get the most enthusiasm, and the decision is made by whichever opportunity happened to arrive on a week when you had cash and energy. A thesis replaces that with a standard: this either fits what I believe or it does not, and if it does not, the fact that it is a good business is somebody else's opportunity, not mine.
The second job is protection against your own narrative. The most expensive mistakes in investing are rarely made by people who had no reason for what they did. They are made by people who had an excellent reason, constructed after the fact, for something they wanted to do anyway. A written thesis with a written falsifier makes that harder, because the reasoning exists in a form that predates the temptation.
The third job is that it tells you what to learn. A thesis has load-bearing assumptions, and those assumptions are a research list. If your belief rests on a market growing at a particular rate, then tracking that rate is your job, and the day it stops growing you have work to do whether or not any position has moved.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Two investors, the same opportunity: a company selling scheduling software to independent dental practices.
The investor with a preference. They like founder-led B2B software and think dentistry is an underserved vertical. The founder is impressive. The deck shows growth. They invest, and their note to themselves says "great team, big market, good product."
Eighteen months later growth has slowed sharply. Was the thesis wrong? There is no way to answer, because nothing was claimed. The investor is now in the position of arguing with a position rather than with a belief, and the available moves are to hold and hope, or to sell and feel foolish. Neither is a decision; both are reactions.
The investor with a thesis. Their written belief is: vertical software wins these practices through the office manager, not the dentist, because the dentist has no time and delegates every tool decision. The companies that scale here will have sales motions aimed at office managers and the strongest signal of that is who attends the demo. I am wrong if the winners in this category over three years turn out to have sold top-down to practice owners or to dental groups.
They ask one question in the first meeting: who attends your demos? The answer is that 70% of demos are booked by owners of multi-site groups. That is not a bad answer - it might be an excellent business - but it is not their thesis. They pass in forty minutes.
Eighteen months later the company has slowed selling into groups, and the investor learns something precise: the group channel is slower than they assumed and the thesis survived a real test. Or the company has grown quickly through groups, and the investor has learned their belief was wrong in a specific way they can now fix. Either way they know something. The first investor still knows nothing except that a position went down.
What it cost to have one. Forty minutes and one written paragraph, against eighteen months of ambiguity.
Believing that a thesis is a summary of what you have already invested in. Written afterwards, from your existing holdings, it is a description with the word thesis on top, and it will never reject anything because it was reverse-engineered from things you already said yes to. A thesis has to exist before the deal it is used on, and the test is simple: has it ever caused you to decline something you found attractive? If not, it is decoration.
Force the causal claim into the open. "Small-business software wins on distribution because the buyer has no procurement process" contains a mechanism; "I like small-business software" does not. If you cannot write the because, you do not yet have a thesis - you have noticed something, which is the honest starting point and a different thing.
Write down what you would have to see to conclude the belief was mistaken, with a date attached. It must be something that could actually happen and that you could actually observe. "The market does not grow" is weak; "this category's three largest by revenue in three years sold top-down" is checkable.
Most theses rest on two or three load-bearing facts about the world - a cost curve, a regulation, a behaviour. List them explicitly, because they are what you now have to monitor, and because it is common to discover while listing them that one is not a fact but a second assumption.
Ask what you know here that a generalist reading the same public sources would not. If the honest answer is nothing, the thesis may still be true, but you have no edge in holding it and should size accordingly, or read more before committing. Enthusiasm is not knowledge and the two feel identical from the inside.
Write a thesis before you begin looking seriously at any category - before the first meeting, not after the third, because after the third you are writing a justification. It is most valuable in private markets, where opportunities arrive one at a time, wrapped in a story, with no price to anchor against and no obvious comparison, so the only standard available is the one you brought with you.
Do not write a thesis for an asset you hold for a structural reason rather than a view. Cash held for near-term spending, a broad index position held precisely because you are not trying to know anything, a house you live in - none of these need a belief about the world, and dressing them in one invites you to start trading them on the strength of a story. A thesis is for the part of a portfolio where you are claiming to know something.
An investor starts with a genuinely narrow thesis - say, infrastructure software sold to engineering teams. An attractive opportunity arrives that is adjacent but not a fit: it sells to data teams rather than engineering. The thesis is amended to "technical teams". Two deals later it becomes "technical buyers", then "companies with a technical sale", and within two years the written thesis admits almost any business-to-business software company. No single amendment was unreasonable and each was made thoughtfully at the time. The cumulative effect is that the standard now matches the portfolio rather than constraining it, and the investor still describes themselves as thesis-driven, accurately by their own records.
Date every amendment to a thesis and keep the old version visible. A standard that is edited to admit the deal in front of you is not a standard, and the drift is invisible unless the versions sit side by side.
Create a free account to read the rest and apply it to your own investing.