How is a transferred call to a human graded?

When a call is transferred from your AI to a person, SuperLedger transcribes the part after the handoff and grades how your team handled it — an A–F letter grade with a score out of 10, plus sub-scores for professionalism, helpfulness, accuracy, communication, and resolution. Severe calls (legal threats, safety issues, cancellation threats) get a Manager Escalation Required banner so the worst calls get reviewed before a client churns over them. You'll find this in the Human Call Analysis card on the call's detail page.

Before you start: open a transferred call from Monitoring → Call Logs (the Human Call Analysis card only appears when a call went to a human).

What the grade shows

  • Letter grade A–F with a score out of 10, in the Human Call Analysis card (marked Post-Transfer).
  • Agent performance sub-scores (each out of 10): Professionalism, Helpfulness, Accuracy, Communication, Resolution.
  • Customer satisfaction — sentiment, frustration and satisfaction signals, and flags like Wait Time Mentioned or Repeated Info.
  • Call timelineAI Duration, Ring / Wait, Human Duration, and Total Call.

Critical flags and escalation

If the call contains something serious, a Manager Escalation Required banner appears with the reason. Critical flags include: Legal Threat, Injury / Safety, Property Damage, Extreme Hostility, Misinformation Given, Unresolved Complaint, and Cancellation Threat.

Turn escalations into training

After a transferred call is analyzed, SuperLedger lists Suggested KB Entries — the questions the human answered that your AI couldn't. Use Copy All to grab them as ready-to-paste Q&A for your knowledge base, so the AI handles them next time.

What happens next

The grade and flags appear once the post-transfer audio is analyzed (recordings can take a few minutes to arrive). The Human Duration is always the total call time minus the AI portion, so the AI leg is never counted against your team.

How it works

Grading runs only on the segment after the handoff, so it measures your team's handling, not the AI's. The sub-scores roll up into the letter grade, while the critical flags are detected separately — a call can be flagged for escalation regardless of its grade.

← Back to Help Center