What i think is for many pipelines or production-scale deployments, you would typically define pipeline-specific variables (or separate variables per workflow) and override them per target (dev, qa, prod). This gives each pipeline its own schedule while still keeping the configuration centralized.