
Classify the exit: thesis, valuation, or risk control
Start by naming what kind of exit this is: (a) thesis invalidation (assumptions broke), (b) valuation/position-size action (it ran ahead of your band)...
Selling decisions break down when every discomfort is treated as the same signal. This checklist helps you classify the exit type—thesis invalidation, valuation/position-size trims, or risk control—then write the exact trigger, evidence, and next review date before you act. Use it after earnings, drawdowns, or major narrative shifts to avoid panic exits, and to avoid “holding a broken thesis” just because the price fell. It also includes a post-exit review loop and re-entry conditions so every sell improves your process instead of rewriting history.

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

Start by naming what kind of exit this is: (a) thesis invalidation (assumptions broke), (b) valuation/position-size action (it ran ahead of your band)...

List the 3–5 assumptions that originally justified owning the business, then ask: which one changed, and is the change durable? Look for fundamental e...

If the thesis is intact but valuation is stretched, prefer pre-written bands: trim part of the position when price implies assumptions you would not u...
Start by naming what kind of exit this is: (a) thesis invalidation (assumptions broke), (b) valuation/position-size action (it ran ahead of your band), or (c) risk control (concentration, leverage, liquidity, or horizon mismatch). If you cannot classify the exit, you will mix rules and sell for the wrong reason.
List the 3–5 assumptions that originally justified owning the business, then ask: which one changed, and is the change durable? Look for fundamental evidence (unit economics, demand, competition, balance sheet, governance), not “the price is down.” If you cannot point to a changed assumption, treat the discomfort as information to review, not as a sell trigger.
If the thesis is intact but valuation is stretched, prefer pre-written bands: trim part of the position when price implies assumptions you would not underwrite today, and exit more only if the risk budget requires it. “All-in/all-out” selling often converts a valuation decision into a narrative bet. Write the rule you used (band, scenario odds, or required margin of safety) and keep it consistent.
Sometimes the business is fine but the position is not: concentration drift, correlation spikes, or rising leverage can make a portfolio fragile. Treat these as portfolio-level exits. Write the risk rule (max position size, drawdown limit tied to liquidity needs, or diversification constraint) and the adjustment (trim, hedge, or reduce exposure elsewhere) so you do not confuse “portfolio hygiene” with “thesis failure.”
Every sell should create a future decision rule. Record what would make you re-enter (evidence threshold, valuation band, or repaired assumption), what would keep you out (persistent thesis break), and when you will review (e.g., next earnings or a quarterly check). Without this loop, sells become one-off emotional relief instead of a reusable process improvement.

Selling because of stress while the core thesis is still intact is one of the most common mistakes. Volatility creates urgency, and urgency creates stories. The checklist forces you to tie the sell to a specific changed assumption or a portfolio risk rule—so you are not just reacting to discomfort.
Sometimes, but many investors do better with tiered trims. If valuation is rich but the business quality remains high, a partial trim can reduce regret and control risk without turning the decision into a binary bet. Full exits make more sense when valuation is extreme and/or the position also violates concentration, liquidity, or thesis rules.
Treat headlines as prompts to re-check assumptions, not as triggers by themselves. Write one or two falsifiable questions (e.g., “Did unit economics deteriorate?” “Did leverage rise meaningfully?”) and look for evidence. If you cannot connect the news to a changed assumption or a risk budget breach, downgrade it to “monitor” rather than “sell.”
Review sells by process quality first. Compare the reason you sold (thesis break, valuation band, or risk rule) with what evidence was available at the time, and whether your documented trigger was actually met. Then record one rule update you will reuse next time, instead of judging yourself only by the subsequent price path.
Re-entry conditions should be evidence-based and time-bound: what must become true again (or what must be proven false) for the thesis to be investable, what valuation band would make the risk/reward acceptable, and when you will check next. This keeps you from revenge-buying a bounce or ignoring a repaired thesis because of ego.
Before the next sell order, classify the exit type and write the exact trigger, evidence, and review date that justify the action.