How should a collection agency prioritise its account inventory?

The short answer

Collector hours are the scarce resource on a collection floor, and a large inventory always has more accounts than hours. Prioritisation decides where those hours go. The sources below agree on the inputs: how likely an account is to pay and how much, its balance and age, how long it has been placed, what the client’s terms allow, and what the law allows. They also agree on the method for changing it: test against a control group before you roll anything out.

What agencies rank accounts on

Propensity to pay. Scoring vendors sell models that rank accounts by their chance of paying. Experian’s Recovery Scores page says they rank charged-off accounts by expected collectability "so your team works the accounts most likely to pay first", and that score bands can guide placement, outsourcing, auction and legal strategy. The same page notes that accounts are often worked by age, balance or last-worked date instead. Experian’s PriorityScore for Collections product sheet describes two outputs: the likelihood a payment is made within six months, and an expected recovery amount. Likelihood and dollars together are what set a sensible order. How a propensity score differs from agent matching.

Balance, age and placement age. A MaxContact article on its 2024 UK benchmarking survey recommends segmenting by age of debt, amount owed and previous payment behaviour, with a strategy for each segment. An Experian piece on agency performance (2012) describes "job sloping": reducing the frequency of collection effort on placements that are less likely to pay, to keep costs in line with what they return.

Liquidation over time. The same Experian piece argues that a blended recovery rate misleads, because it mixes fresh and old placements. It recommends tracking each month’s placements as a vintage and plotting its liquidation curve month by month. For prioritisation, the curve shows when a batch of placements tends to pay, and when extra effort on it stops returning much.

Client terms. A Kaulkin Ginsberg best-practice piece for recovery managers on insideARM (2009) describes credit issuers recalling accounts that have not paid after a set period and placing them with a competing agency. A recall date is a deadline on the inventory: an account due to go back to the client has to be worked before it leaves.

The statute of limitations and contact rules. Under Regulation F section 1006.26, a debt collector "must not bring or threaten to bring a legal action against a consumer to collect a time-barred debt", which changes what an aged account can be worked toward. Section 1006.14 presumes a debt collector complies with the call-frequency limit if it calls a person about a particular debt no more than seven times within seven consecutive days, and not within seven consecutive days after a telephone conversation about that debt, subject to the exclusions in the rule. Calls are a fixed budget per account per week, so they are best spent on the accounts that matter most. Your compliance team and counsel decide how these rules apply, along with state rules and client requirements.

Work strategies and champion/challenger testing

A work strategy turns the ranking into action: which accounts go to the dialer today, how often each is attempted, which get a letter, a text or an email, and which wait. The standard way to change one is champion/challenger testing. A 2006 insideARM tip by Stephen Leonard describes it as selecting two statistically random sets of accounts and treating the control (champion) and test (challenger) groups differently, so that any difference in results comes from the different actions. For letter tests it recommends a minimum period, typically twelve months, before drawing conclusions.

The same idea is used to compare agencies. An Experian piece (2014) says results should lead to "the allocation of more accounts of similar types to the champion, less to the challenger", and an IC System article (2023) describes splitting a portfolio between competing agencies and shifting share to the top performer on a quarterly scorecard.

Channel mix

Phone is not the only lever. The MaxContact article recommends a mix of calls, email, SMS and other channels to improve right-party contact. Each digital channel has its own rules: under Regulation F section 1006.6, an email or text from a debt collector must include a clear and conspicuous statement of a reasonable and simple way to opt out of further electronic communications to that address or number. In our view, channel choice belongs inside the work strategy and gets tested the same way.

How to tell whether a prioritisation change worked

Split accounts at random before any work starts, run the new strategy on one group and the current one on the other in the same weeks, and compare dollars collected per account contacted. Comparing this quarter with last quarter mixes in new placements, the season and staffing. How to prove a change worked sets out the method.

The dimension most strategies leave out

Every input above describes the account. In our view, most prioritisation treats the collector as interchangeable: once an account is chosen for today, it goes to the next free collector, or to whoever tops the leaderboard. But collectors convert differently on different kinds of account: one does better with large, complex balances, another with consumers who need time and patience, another clears small balances quickly. A single leaderboard averages those strengths away, and the order in which accounts are worked says nothing about who should work them. Why collectors convert at different rates and why your best collectors cannot take every account go further.

Where agent matching fits

Agent matching (predictive pairing) works on that last decision. The agency keeps its scores, its work strategy, its order and its rules; BestPair decides which collector works each account the agency has chosen, from the floor’s own call and payment history. BestPair never contacts consumers and does not change offers, scripts, contact rules or call frequency. Whether that decision is worth anything on a given floor is what the Floor Assessment measures from the agency’s own history, and live pairing is measured against a control group, the same way as any champion/challenger test.

Sources

• Experian Recovery Scores: Experian product page on ranking charged-off accounts by expected collectability; a vendor’s description.
• Best practices for improving agency performance: Experian Insights blog, guest contributor, on vintage liquidation curves and job sloping.
• Can you compare the performance of your collection agencies?: Experian Insights blog, guest contributor, on champion/challenger allocation between agencies.
• PriorityScore for Collections: Experian product sheet on a likelihood-of-payment model and an expected-dollar model; a vendor’s description.
• Optimising your debt collection contact centre performance: MaxContact article on its UK benchmarking survey, with recommendations on segmentation and channels.
• Recovery management: best practices for managing collection agencies: Kaulkin Ginsberg on insideARM, on placement, recall and agency specialisation.
• Champion-challenging letter strategies, part one: Stephen Leonard on insideARM, on random champion and challenger groups.
• Champion-challenger collections model: IC System, Eric Johannes, on testing agencies against each other on the same debt.
• Regulation F, harassing, oppressive or abusive conduct: the CFPB’s text, including the call-frequency presumptions.
• Regulation F, time-barred debts: the CFPB’s text on legal action on time-barred debt.
• Regulation F, communications in connection with debt collection: the CFPB’s text, including the opt-out notice for email and text.