How do you calculate RPC rate, promise-to-pay rate and kept rate?

The five numbers, as formulas in words

RPC conversion is the measure used across these pages: the share of right-party contacts that end in a payment or an arrangement. What RPC conversion is, and how to measure it.

Pick each denominator once

Most arguments about these numbers are arguments about the bottom of the fraction.

• RPC rate per attempt or per account. Per dial attempt describes the dialer and the data. Per account worked describes how much of the placement the floor reached. Both are fair; mixing them month to month is not. insideARM’s Ask the Experts column (2008), answered by Kaulkin Ginsberg, treated right-party contacts per agent per hour as the measure of dialer productivity, and dollars collected and promises as the measure of quality. That split still holds: contacts are the dialer’s number, conversion is the floor’s.
• Promise-to-pay rate per right-party contact. Dividing promises by all calls, or by all accounts, mixes contact into conversion. If the dialer reaches more people, a per-call promise rate rises without any collector doing better.
• Kept rate per promise falling due. Promises taken in the last week of a month are still open at the month end. Divide kept promises by promises made in the month and the late ones count as failures. Divide by promises whose date has passed, and wait out the grace period your floor allows before you count.
• Dollars per account contacted, as payments land. Count payments when they reach your collection system, and take reversals back out.

Why kept rate and dollars count more than promises

A promise to pay is an outcome code a collector enters when the call ends. A kept promise is money in the account. The gap between them is where a floor loses track of what it collects.

A floor that rewards promises gets more promises, including ones the consumer was never going to keep. An AccountsRecovery.net webinar for agency executives (2020) was built on that worry: a promise is only as good as the word of the person giving it. In our view, a collector who agrees a smaller amount the consumer can actually pay often brings in more than one who takes a larger promise that breaks.

Even kept rate has a blind spot. It counts every kept promise as one, whatever its size. That is why dollars collected per account contacted is the score we use for any change in routing: it weighs how often a contact pays and how much. What counts as success in collections.

What published figures can and cannot tell you

Benchmarks are rare, and their definitions are rarer. MaxContact, a contact centre software company, published a 2024 UK benchmarking survey of 100 debt collection professionals: the mean RPC rate reported was 26% and the mean promise-to-pay rate 29%. The article does not define the denominator for either rate, so neither can be set against your own numbers. They describe the teams who answered, in the UK, in that year. Sales floors meet the same problem with outbound close rates.

In our view the more useful comparison is inside your own floor: the same measure, the same denominator, by type of account and by collector. The spread between your own collectors on the same kinds of accounts says more than any outside figure. Why collectors convert at different rates, and what a fair collector scorecard measures.

Sources

• insideARM, Ask the Experts: how can a collection agency best measure dialer productivity? A 2008 trade press column answered by Kaulkin Ginsberg on contacts per hour and quality measures.
• MaxContact, 2024 UK benchmarking insights A software vendor’s survey of 100 UK debt collection professionals.
• AccountsRecovery.net webinar: keeping collectors from buying into the promise of false hope A 2020 trade webinar for agency executives on promises that do not hold.