a builder's codex
codex · patterns · Subtraction beats addition — Taleb, Munger, and Hormozi on why removal is more reliable than improvement

Subtraction beats addition — Taleb, Munger, and Hormozi on why removal is more reliable than improvement

Convergence

Three operators from risk theory (Taleb), investing wisdom (Munger), and offer engineering (Hormozi) converge on the same default heuristic: in complex systems with poorly-understood dynamics, removal is more reliable than improvement, and inversion ("what would guarantee failure?") beats invention ("how do I succeed?"). Each operator applies the principle to a different layer — products, decisions, offers — but the underlying discipline rhymes: subtract first, add only when forced.

Operators

Variation

The three operate at compounding scales: cognitive (Munger), operational (Hormozi), systemic (Taleb). Together they argue that the default move in any improvement context should be subtractive, with additive interventions reserved for situations where the team has high confidence in the diagnosis and the cost of being wrong is bounded.

Implication

For founders, PMM leads, product managers, and operators reviewing stuck systems:

1. Quarterly subtraction reviews. Alongside the "what should we start" exercise, run a "what should we stop" exercise — and weight the stop list at least equally.

2. Inversion before invention. When facing a hard decision, ask "what would guarantee failure here?" and design to avoid those conditions before designing to seek the success path.

3. Iatrogenic restraint on big changes. Before any aggressive intervention (org restructure, big rewrite, sweeping policy), ask: is the proposed cure more dangerous than the disease? Default to the smallest reversible move that addresses the real problem.

4. Subtractive growth audit. When growth stalls, the first questions are about removal: which channel is masking weakness, which audience is mismatched, which offer component is dragging conversion. Add only after the subtractive review has cleared.

Counter-evidence

Sources

See also Decision quality at scale comes from process design, not from individual brilliance or harder thinking for the broader Kahneman + Munger decision-quality pattern.

Open the interactive view →