Link partner records

Governance before technical validation

Before architecture review starts, the internal CRM record, the hyperscaler opportunity record, and — when Databricks is in the workload — the Databricks partner-portal record all have to reference the same customer, workload, and identifiers. This is what makes usage and attribution reconcilable at go-live instead of becoming a support ticket six months later.

When a single workload runs both a Microsoft or Google Cloud consumption path and Databricks consumption in parallel, the two records stay linked but distinct — one workload can still produce separate, correctly attributed usage meters on each platform.

Source classification matters here

Each linked record carries a source tag — partner-created, field-sourced, or partner-influenced — because that classification is what downstream attribution and funding-eligibility logic reads from later, not the account team's memory of how the deal started.