
Qualify ideas before they become candidates
Treat “interesting” as a rejection state. Only promote an idea to a candidate if you can write: what must be true, what would change your mind, and wh...
A watchlist is a list of hypotheses, not a portfolio. This toolkit gives you a simple gate for turning an idea into a position: qualify evidence and timeframe, write a one-sentence thesis, define 2–3 invalidation signals, set a sizing cap and risk boundary, and schedule the first review date. The goal is fewer impulse entries and a decision trail you can audit later. Use it for long-term or swing candidates; if you want “signals” or guaranteed entries/exits, this is not that, and it does not replace fundamental research or valuation work.

Pick the smallest next action now: test your bias pattern, run a scenario, or copy a prompt before making a portfolio move.

Treat “interesting” as a rejection state. Only promote an idea to a candidate if you can write: what must be true, what would change your mind, and wh...

Summarize the thesis in one sentence, then list 2–3 disconfirming signals that would make you pause or exit. If you cannot name what would invalidate...

Decide the position’s role (core/learning/hedge), the maximum size you are willing to reach, and the risk boundary you will respect if the thesis weak...
Treat “interesting” as a rejection state. Only promote an idea to a candidate if you can write: what must be true, what would change your mind, and why now is a reasonable window to evaluate. This prevents watchlists from becoming entertainment.
Summarize the thesis in one sentence, then list 2–3 disconfirming signals that would make you pause or exit. If you cannot name what would invalidate the idea, you do not have a thesis—you have a story, and stories create impulse entries.
Decide the position’s role (core/learning/hedge), the maximum size you are willing to reach, and the risk boundary you will respect if the thesis weakens. The goal is not prediction—it is preventing “accidental concentration” and reactive averaging.
Before executing, capture the minimum checklist: entry condition, evidence you relied on, the first review date, and the single most important metric or signal you will monitor. You are creating a future review trail, not optimizing a perfect entry.
Track which candidates became positions and why. If many names “graduate” without strong evidence, tighten the gate. If none ever convert, your filters may be unrealistic. The aim is better decision quality and fewer noisy actions, not more trades.

Because “interest → order” is where emotion sneaks in. A workflow forces each idea through the same gate: thesis, invalidation, sizing boundary, and review plan. That consistency reduces impulse entries and makes your decisions auditable later.
Only a small fraction. A good workflow is designed to reject most ideas quickly and cheaply, then allocate deeper work to the top few candidates. If everything converts, the gate is too loose and you are likely over-trading or over-confident.
Remove or archive it when the thesis is invalidated, the key evidence never appears by your review deadline, or the attention cost is no longer justified. Stale names are dangerous because they keep you “half-committed” and increase impulse decisions.
No. It organizes the work so you do not skip it. Your research (fundamentals, valuation context, risks) should feed the thesis and the “no” conditions, and the workflow makes sure you document what you relied on before any execution.
Yes. Start with the smallest version: one-sentence thesis, one invalidation trigger, one sizing cap, and one review date. The purpose is calm structure—fewer decisions, clearer boundaries, and a learning loop—not perfection.
Take your top three watchlist names and force them through one qualification gate before any new capital is deployed.