Thesis Drift

The gradual widening of a written thesis, one reasonable amendment at a time, until it describes the portfolio rather than constraining it.

Example

A thesis about infrastructure software sold to engineering teams becomes "technical teams", then "technical buyers", then "companies with a technical sale" - each change made thoughtfully, admitting almost anything after two years.

Important

No single amendment looks wrong, which is why it is invisible without dated versions kept side by side. Never edit a thesis in place.

Related terms

  • Investment thesis — A written statement of what you believe about an investment, at what price it is worth owning, and what specific observation would tell you that you were wrong.
  • Load-Bearing Assumption — One of the two or three facts about the world that a thesis rests on, such that if it stopped being true the whole belief would fail.
  • Hindsight bias — The tendency, once an outcome is known, to recall having expected it - the genuinely uncertain version of events quietly stops being available.

Lessons that explain it

  • Ask whether it fits your thesis before asking whether it is good — A good company outside your thesis is a different decision from a good company inside it. Checking fit first stops attractive deals from quietly rewriting your strategy. (9 min)
  • If you cannot write a good bear case, you do not understand the deal — A bear case built from generic risk categories teaches nothing. A bear case built from 'what would have to be true' names the specific number the whole thesis depends on. (12 min)
  • Amend, date and keep the original - never quietly rewrite — A memo is a record of reasoning at a moment. When facts change, add a dated amendment rather than editing the original, so the evolution of your thinking stays visible and reviewable. (10 min)
  • The annual hour that decides whether you learn anything — A thesis that is never reviewed is a belief you happen to have held for years. The review is where a written thesis converts into judgement, and it has to separate what you got right from what merely worked. (11 min)

All 578 terms in the Investing glossary