One Variable

"One variable" thinking is the discipline of changing or analysing one factor at a time while holding others constant — the experimental and mental equivalent of ceteris paribus. In complex systems, outcomes depend on many factors; isolating one lets you infer cause and effect and avoid confounding. Founders use it in experiments (A/B tests), strategy (what if we only changed price?), and diagnosis (why did conversion drop?). The lens: before concluding "X caused Y," ask whether you’ve held other drivers fixed or varied them; if not, your inference is weak.

Design tests and analyses so that only one (or a small set of) factor(s) vary. When interpreting results, list what was held constant and what wasn’t; if many things changed, treat causation as tentative. In strategy, use one-variable counterfactuals to prioritise: "If we could only fix one thing, which would move the needle most?"

One variable: change or analyse one factor while holding others constant. Use it in experiments, strategy, and diagnosis to infer cause and avoid confounding. Founders apply it in A/B tests, prioritisation, and root-cause analysis.