How BlueBear handles the work
Use accepted outcomes as the denominator
A completed model call is not necessarily a valid business result. Reconcile accepted outcomes with the workflow owner or system of record.
Evidence: Attempt count, accepted outcome count, rejection reasons, and reconciliation period
Include recovery and human work
Retries, fallbacks, approvals, queue handling, and exception resolution can materially change unit economics even when provider cost looks low.
Evidence: Correlated retry events, active review time, loaded labor rate, and exception time
Document allocation assumptions
Use an approved, repeatable method to allocate runtime and shared tool costs. Keep the measurement window, rate source, exclusions, and workflow version beside the result.
Evidence: Reproducible input sheet and dated methodology
Compare policy changes, not vanity totals
Measure the same workflow before and after a routing, prompt, tool, or approval-policy change. Track cost, acceptance, latency, and failure rate together.
Evidence: Comparable versions, sample sizes, quality gates, and confidence range