Making the company's own evidence outrank generic advice

• we tried that
• again
• already did
• lesson
• learned
• repeat

• What has this company already tested, and what did it find?
• Is the result written where the next decision will see it?
• Does any recorded lesson contradict the general advice being followed?
• How confident is each lesson, and on what sample?

• recorded lessons with results
• the hypothesis each tested
• the sample and window

• Results not written down
• Lessons written but never retrieved

• If
  A recorded lesson contradicts general advice
  Then
  The company's own verified evidence wins, and the plan says so out loud.
• If
  A lesson rests on a tiny sample
  Then
  Hold it as weak evidence, not as law.
• If
  Nothing is recorded
  Then
  The first improvement available is recording results at all.

After the next completed test, write the hypothesis, the result, the verdict and whether to repeat, and check it surfaces in the following decision.

Read after 30 days.

• The lesson is retrievable
• The next related plan visibly reflects it

• Lessons are written but never influence a plan — the retrieval is broken, not the writing

• A lesson is retrieved and used → The loop works; keep the discipline.
• It is not → Fix retrieval before writing more.

Aury can carry out: read company, file memory, prepare document. Specialists involved: strategy.

Applies always, and is the reason the rest of the corpus improves over time.

A lesson from one company at one moment does not generalise to others, and Aurygine keeps it strictly inside that company.

• Aurygine synthesis: Aurygine synthesis across founder journeys

Confidence: high. Last reviewed 2026-08-25. Your company's own recorded results override this playbook whenever the two disagree.