Launch materials reviewed by a specialist panel

The decision tree

5 typed decisions, each with an action for every answer

At fixed moments in a run, the engine puts one narrow question to a decision model. The decision model never writes text: it answers yes or no with a probability, picks from listed options, or gives a score, about a small slice of the material. The engine then does exactly what this tree says, which is what makes the run auditable. The thresholds are the template's design values, not measured results.

1. 1

   Before workers, before a task runs

   Small worker or strong modelYes or no, with a probability

   Before a task runs

   Does this launch piece state prices, plan names, limits or discounts, so that the pricing reviewer must read it?

   Sees only: One launch document's headings and any figures in it

   Why: Each reviewer reads only the pieces its reference document can judge.

   00.20 and 0.501

   • Yes: 0.50 or higher→thenAccept
   • Unsure: 0.20 up to 0.50→thenAccept
   • No: below 0.20→thenSkip this unit
2. 2

   Small worker or strong modelYes or no, with a probability

   Before a task runs

   Does this passage make a performance, security, compliance or comparative claim that needs the strong claims reviewer?

   Sees only: One passage of launch copy

   Why: The claims that attract legal trouble get the most careful reader.

   00.25 and 0.501

   • Yes: 0.50 or higher→thenEscalate to a strong model
   • Unsure: 0.25 up to 0.50→thenEscalate to a strong model
   • No: below 0.25→thenAccept
3. 3

   After workers, the judge checks

   Evidence checkYes or no, with a probability

   After a worker answers

   Does the quoted copy promise a capability or figure that the quoted specification line does not state, or states only with a condition the copy drops?

   Sees only: The quoted copy and the quoted specification line

   Why: Catches the dropped condition, which is how accurate specifications become inaccurate launches.

   00.50 and 0.851

   • Yes: 0.85 or higher→thenAccept
   • Unsure: 0.50 up to 0.85→thenEscalate to a strong model
   • No: below 0.50→thenReject and retry
4. 4

   Reconciler, while merging

   Conflict checkYes or no, with a probability

   While reconciling

   Do these two launch documents give different values for the same plan name, price, limit or availability date?

   Sees only: Two quoted passages from different launch documents

   Why: Each owner reviews only their own piece, so cross-document drift is nobody's job.

   00.50 and 0.851

   • Yes: 0.85 or higher→thenAccept
   • Unsure: 0.50 up to 0.85→thenEscalate to a strong model
   • No: below 0.50→thenContinue
5. 5

   Accountable person, before anything is settled

   Person decidesA choice among options

   Before anything is reported as settled

   Which kind of finding is this, and therefore whose sign-off does it need before launch?

   Sees only: One verified finding with its brief

   Why: Legal, product and pricing each own their findings, and the launch lead decides go or no-go.

   Accountable: Legal owns claims, product owns technical accuracy, and the launch lead decides go or no-go.

   • Unsupported or unlisted claim→thenAsk a person
   • Technical error→thenAsk a person
   • Pricing inconsistency→thenAsk a person
   • Brand voice note only→thenAccept