How are collection calls monitored, scored and root-caused?

The rest of this page shows who listens, how calls are picked, what the scorecard does and how a pattern becomes a fix. Timing and sample sizes can change by contract.

Who listens, and who they report to

The scoring and audit groups sit outside operations on purpose, so the people judging a call don't answer to the people measured on collections. TSI's Chief Compliance Officer is independent of operations and reports to the CEO and to the Risk and Compliance Committee of the Board.

What gets recorded, and for how long

Every collection call is recorded. A log of recorded calls lets staff pull a call by consumer, collector, call type, date or time, and access is limited to authorized supervisory, compliance and quality staff.

Clients can get the recordings for their own portfolios through secure remote access, SFTP extracts or a secure API. Recordings are kept for the period the contract sets, and never less than the law requires: Regulation F says a collector that records calls "must retain the recording of each such telephone call for three years after the date of the call" (12 CFR 1006.100(b)).

How calls are chosen for review

Supervisors mostly pick calls at random. They pick specific calls when a complaint names a collector or when a call runs unusually long.

Central monitoring starts each month from the staff roster. Every collector scheduled to work is assigned to a quality coordinator, who scores about 5 of that collector's calls in the month, and more when there's a known issue.

Ripple helps choose those calls. It scores every call each day and flags calls in risk areas: bankruptcy, cease-and-desist requests, attorney involvement, callback requests, disputes, voicemail handling, and rude or unprofessional behavior from either side. It also tracks whether the call-recording disclosure, the collector and company identification and the required collection disclosures were given.

Flagged calls go to a person. A quality coordinator listens and scores each one, and the software's flag doesn't count against a collector on its own. Ripple doesn't take account actions or decide outcomes.

The scorecard

Each reviewed call is scored against a list of compliance attributes, each marked pass, fail or not applicable, plus a quality attribute for how the call was handled.

The compliance attributes follow the rules a call has to meet. One is the disclosure federal law requires: that the collector "is attempting to collect a debt and that any information obtained will be used for that purpose," and in later calls that the communication "is from a debt collector" (15 U.S.C. 1692e(11)). Others cover identifying the right party, handling a dispute or a cease request, and stopping when an attorney is involved.

A fail on any compliance attribute gives the collector 1 compliance point. A call with several fails still counts as 1 point, and repeat fails on the same attribute on the same call date count once.

What happens after a failed call

The supervisor delivers the result. The collector reviews it and acknowledges it within 72 hours, and a tracking report shows every acknowledgement that's still open.

Consequences build from there:

• Supervisors coach one-on-one during the month, and each collector signs a month-end scorecard showing the rating and what to change.
• Compliance results are built into monthly incentive pay. A collector with a compliance point takes a reduction.
• Repeat compliance points lead to progressive discipline, and rude or unprofessional conduct goes straight to it.

From patterns to root causes

Single calls get coached. Patterns get fixed, and that work follows a set path:

1. A daily report shows fails by attribute, collector and team, and a daily complaint report lists every complaint received the day before.
2. The quality manager reviews the trends and recommends corrective action or a policy change.
3. Complaints from any source, including the CFPB, the Better Business Bureau and state attorneys general, go to Consumer Affairs. Collection pauses on the account while the team investigates and finds the root cause.
4. Each finding becomes a corrective action plan with an owner, a timeline and success criteria.
5. The fix is tested after it's made, and the plan closes only when follow-up monitoring or an audit shows it worked.
6. When a fix changes a procedure, operations and compliance leaders review it, the responsible vice president signs off, and the client approves it before it goes into the collector knowledge base and training.

Senior management reviews complaints monthly for trends. Findings and remediation go to the Management Compliance Committee and the Board's Risk and Compliance Committee. The complaint process itself is described in how TSI handles complaints.

What a client can see and join

Clients can listen to recordings for their portfolios, ask for quality results for the collectors on their program, and receive complaint reports. They can also join calibration sessions with supervisors and quality staff, monthly or every other week, where everyone scores the same calls so a "pass" means the same thing to both sides.

Call limits have their own rule brief: Reg F call limits: what counts toward 7-in-7.

Questions to ask any agency

• Which calls are recorded, and for how long are recordings kept?
• How many of each collector's calls does a person score each month, and how are they picked?
• Who does the scoring report to?
• What happens to a collector's pay and standing after a compliance fail?
• Show us one root-cause finding from the last year and the test that closed it.

The full set, with weights and a scoring template, is in what a collection services RFP should include, and how to score it. How accounts move from placement to close is on its own page.