Ideas that sharpen your research process
Most private investors spend the majority of their research time on individual positions: reading reports, studying competitive dynamics, forming a view on management, and stress-testing a valuation. That work is genuinely valuable, but it addresses only one dimension of the research problem. A collection of well-researched positions can still form a poorly constructed portfolio if the positions share hidden assumptions that have never been made explicit. The issue is not whether each idea is good on its own terms. The issue is whether the ideas, taken together, are all quietly betting on the same thing. Two companies in entirely different sectors can both depend, in ways that are not immediately obvious, on a continuation of low borrowing costs, or on a particular pattern of consumer spending holding steady, or on a regulatory environment remaining benign. When the underlying condition they share comes under pressure, positions that appeared diversified begin to move in the same direction at the same time. That is not diversification failing; it is the absence of portfolio-level thinking making itself felt.
One useful discipline is to write down, for each position, not just the investment thesis but the two or three conditions that must remain broadly true for that thesis to play out over time. These are not predictions. They are the load-bearing assumptions embedded in your reasoning. Once you have done this for every position, you can lay those assumption sets side by side and look for repetition. You may find that a surprising number of your ideas rest on the same macro condition, the same sector tailwind, or the same view about how a particular kind of business model will be valued by the market. When you see that repetition, you are not necessarily looking at a mistake. You may have genuine conviction in that shared assumption. But you are now looking at a concentration of risk that deserves to be acknowledged and sized accordingly, rather than one that remains invisible because it is spread across positions that appear, on the surface, to be unrelated.
Scenario thinking is a practical tool for surfacing these connections before they become expensive. Rather than asking what happens to each position in isolation, ask what happens to your entire portfolio under a small number of plausible but distinct scenarios. A scenario in which growth expectations are revised downward across a broad range of industries will affect your positions differently from a scenario in which a specific sector faces sudden regulatory scrutiny, and both of those will look different from a scenario in which the cost of capital rises sharply over a short period. Working through these scenarios does not require precise forecasting. The point is not to predict which scenario will occur but to understand which of your positions would be most affected, which would be least affected, and whether the portfolio as a whole would remain coherent under conditions that differ from the ones you implicitly assumed when you built it. If nearly every position suffers under one particular scenario, that scenario deserves your attention regardless of how unlikely you currently believe it to be.
There is also a subtler version of this problem that has less to do with macro conditions and more to do with the internal logic of your research process. If you tend to favour a particular style of analysis, or to be drawn to a particular kind of business, or to weight certain types of evidence more heavily than others, then your portfolio may reflect not just your considered views about the world but also the consistent biases in how you gather and interpret information. A portfolio built entirely from one type of source, or one analytical framework, or one mental model of how value is created, is exposed to the risk that the framework itself is wrong in ways you have not yet discovered. This is not an argument against having a consistent investment philosophy. It is an argument for being honest about what that philosophy assumes, and for occasionally testing your conclusions against perspectives that do not share your starting point. A coherent research portfolio is one in which the relationships between positions have been examined, not just the positions themselves.