Can anyone explain why the Quality Manager score for a specific agent diverges from the WFM forecasted adherence by 15%? The Architect flow is configured to trigger evaluations only after successful disposition, yet the WFM dashboard reflects these interactions as unaccounted time.