Fixture, campaign, consent, paid value and scan state.
Prove the data path before a club shares real fan data.
This rehearsal exercises the same ticket-grain logic used by the pilot, but only with the repository’s synthetic Brighton dataset. It proves the workflow, not a London City commercial outcome.
Pseudonymous matching is used only to calculate the cohort.
Only counts, rates and economics are needed by the product layer.
Attribution is visible; incrementality remains unproven.
The same decision ingredients the live pilot will need.
Consented opener buyers who have not purchased Brighton.
Descriptive attribution from the synthetic current-fixture records.
Observed in the rehearsal cohort; not an incremental-lift estimate.
Synthetic realised ticket value.
Based on records with a decided scan state.
The production bridge persists aggregate evidence only.
What this proves, and what it does not.
The workflow can turn ticket-grain input into repeat, attribution, revenue and attendance-quality evidence.
The product can work with aggregate club evidence without storing supporter-level records in the repository.
A real pilot still needs authorised club data and a pre-agreed counterfactual if the club wants a causal lift claim.
Replace the synthetic input, not the product workflow.
A pilot club supplies an authorised export outside the repository. The same aggregation contract then feeds Opportunity, Impact and Results with measured evidence.