The Discipline · No. 01
Evidence before scale
The one rule that reorders every other rule in the firm.
By Bruno Hounkpati · ≈1 min read · Published July 2026
Hypothesis
A decision that has not been tested against its own evidence is a decision that has been outsourced to eloquence. Scale multiplies whichever we choose.
Evidence
- Every stalled programme we inherit shares one trait: the deciding memo cites conviction rather than measurement. The pattern holds across pricing moves, restructurings and AI pilots we have re-opened.
- When the same programme is re-run with a written hypothesis and a pre-committed verdict standard, the argument shortens and the meeting ends with a call, not a follow-up. We have observed this on both sides of the corridor, in French and English rooms.
- The cost of shipping an untested claim is not the money spent — it is the six months the organisation cannot re-use. Reversing an announced decision is dearer than testing it once, and the ledger is where that cost lands.
Verdict
If you cannot state the hypothesis, the evidence, and the verdict standard on one page, do not scale it. Test it first — for ninety minutes if that is all you have.
Citations
- Method statement — Stredge Partners · The 90-minute scan · v2026.2 (available on request before an engagement letter).
- Internal review — Stredge Partners casework, 2024–2026. Individual engagements are confidential; the pattern is not.
Take this to your desk