Collections call QA scorecard: what to score and how
A practical guide to building a collections call scorecard: the sections, how to weight them, and how to keep reviewers consistent.
What should a collections call scorecard include?
A good collections scorecard has five to seven sections, each with two to four yes/no or 1 to 5 questions. Keep it short enough that a reviewer can score a call in one pass. Long scorecards produce rushed scoring and uneven results.
Step 1: Separate conduct from outcome
Score how the agent behaved (tone, language, accuracy) separately from what the call achieved (promise to pay, payment, dispute noted). An agent can behave well and get no promise, or get a promise by behaving badly. One combined score hides both.
Step 2: Mark some items as must-pass
Some items should fail the whole call if missed, for example abusive language or a false statement about consequences. Decide these with your compliance and legal team. Do not copy a list from the internet.
Step 3: Handle calls where an item does not apply
A very short call, a wrong-number call or a call that was cut off cannot be scored on every item. Add a "not applicable" option and exclude those items from the total. Otherwise short calls score full marks or zero for the wrong reasons.
Step 4: Weight what matters most to your business
Give more weight to the items that cost you most when missed. Write the weights down so the scorecard can be explained to agents.
Step 5: Calibrate reviewers
Once a month, have every reviewer score the same five calls and compare. Where scores differ by more than one point, discuss the wording of the question and fix the wording.
Where does Cadence fit?
Cadence is a call quality assurance tool for collections teams at NBFCs, banks, lenders and BPOs. It scores every recorded recovery call against a collections-specific scorecard, with timestamped reasoning, and turns the result into coaching and per-call incentives. Team leads can override a score. Learn more on the Cadence site.