Enterprise debt recovery KPI framework for platform evaluation

TL;DR

• Score cadence and compliance enforcement by testing whether each platform blocks prohibited calls and applies post-conversation cooldowns under Regulation F.
• Test auditability through call-level and debt-level records that support regulatory examinations and litigation discovery.
• Evaluate multi-state configurability by checking whether state disclosure rules trigger the applicable voicemail controls and flag conflicts with federal limited-content message requirements.
• Compare litigation and escalation capability after amicable outreach fails, including support for small claims, liens, filing, and court processes.
• Assess onboarding and intake quality through documented handoffs, data validation, and exception handling.
• Request recovery-rate and cost-per-dollar-collected figures separately. Require each vendor to disclose its calculation method, measurement period, account cohort, exclusions, denominator, and sample size. Any comparison with ACA International or Federal Reserve data must cite a specific dataset that measures a comparable population and outcome.

Why generic vendor comparisons fail enterprise buyers

Generic vendor comparisons often emphasize interface quality, integrations, and feature counts. Those criteria help assess usability, but they reveal little about whether a platform controls regulatory risk across large account volumes. A configuration error can apply the wrong communication rule across every account governed by that setting before manual review catches it.

Regulation F provides a concrete system test. A collector is presumed to violate the rule when it places more than seven calls within seven consecutive calendar days to a person about a particular debt. The same presumption applies when the collector calls within seven calendar days after a telephone conversation about that debt, according to the CFPB Debt Collection Rule FAQs.

An enterprise buyer should test whether a platform blocks the eighth call automatically and starts the post-conversation cooldown without manual intervention. Buyers should also inspect whether those controls operate per person and per debt. A policy document or sales-page compliance claim cannot prove either behavior.

A useful comparison therefore scores observable controls and the evidence behind them. Buyers can place these criteria directly into RFPs, vendor demonstrations, and pilot acceptance tests to compare enforcement mechanics rather than marketing claims.

The enterprise debt recovery KPI scorecard

Apply a four-point evidence scale to every category. Score 0 when a vendor offers only a sales claim. Score 1 when documentation describes the control. Score 2 when a live demo proves the control works. Score 3 when a pilot produces exportable evidence under realistic volume. A zero in a mandatory compliance category should disqualify the platform rather than disappear inside an average score.

1. Automated call-cadence enforcement

Require the platform to count calls per person and per debt across a rolling seven-day window. Under Regulation F, placing more than seven calls within seven consecutive calendar days creates a presumption of a violation. A score of 3 requires the platform to block the eighth call automatically and record why it blocked the attempt.

2. Conversation-triggered cooldown logic

Require a completed telephone conversation to start a separate seven-day cooldown for the same person and debt. The vendor should demonstrate how its platform identifies a qualifying conversation, suppresses later calls, and handles agents working across different queues. Policy reminders and agent warnings deserve lower scores than an enforced block.

3. Voicemail and limited-content message control

Require authorized compliance staff to lock voicemail templates to the content permitted for a limited-content message under Regulation F. The message must include a business name that does not indicate debt collection, a request that the consumer reply, the name or names of one or more natural persons whom the consumer can contact, and one or more telephone numbers for the reply. The platform should prevent agents and automated tools from adding content that would make the message ineligible for limited-content message treatment.

Use Zortman v. J.C. Christensen & Associates, Inc., 870 F. Supp. 2d 694 (D. Minn. 2012), as historical context rather than as the source of the current federal limited-content message rule. For a current negative test, ask the vendor to enter wording such as “This is a call from a debt collector.” Template controls should flag or reject that wording because a limited-content message may not use a business name or additional content that indicates debt collection.

4. Multi-state rule mapping

Require the platform to connect each account with the applicable federal and state communication rules. A state-mandated voicemail disclosure may add content that falls outside Regulation F’s definition of a limited-content message, so the vendor must determine whether a national template remains eligible in each applicable state. A high-scoring platform identifies the conflict before use and preserves the rule version behind its decision.

5. Audit-log production

Require call-level and debt-level records that reconstruct each compliance decision. Logs should connect every attempt and conversation to the applicable counter. They should also preserve blocked actions and the script version used. During a CFPB examination or litigation discovery request, the vendor should export those records without relying on manual reconstruction.

Recovery rate, cost per dollar collected, and right-party-contact rate belong in the commercial portion of the scorecard, but the sources here do not provide defensible benchmarks. Request each metric directly from vendors with its calculation method attached. The methodology should define the measurement period and account cohort, then disclose exclusions and the denominator. Buyers should not convert unsupported marketing claims into scorecard benchmarks.

Compliance enforcement mechanics to test in a vendor demo

Run each compliance scenario in a sandbox and require the platform to produce a clear pass or fail through its behavior. A compliance officer’s description of company policy cannot establish whether the software enforces that policy.

• Eighth-call block. Load seven call attempts for one person and one debt within a rolling seven-day period, then attempt another call. The platform should block the eighth call before dialing and record the applicable rule. Regulation F creates a presumption of violation when a collector places more than seven calls within seven consecutive calendar days.

• Post-conversation cooldown. Record a live telephone conversation about the debt, then schedule another call before seven calendar days have passed. The platform should prevent the call automatically rather than depend on an agent checking notes.

• Counter scope. Create two debts for one person and place calls against each debt. The platform should calculate frequency using the correct person and debt records rather than relying on a broad account-level counter.

• Voicemail controls. Ask an administrator to lock a limited-content message template to permitted content, then have an agent try to add collection language. The platform should reject unauthorized changes. Next, apply a state-specific disclosure and confirm that the platform identifies whether the added wording falls outside Regulation F’s permitted limited-content message content.

• Audit-log production. Request an export showing each call attempt, conversation, blocked action, voicemail version, rule configuration, and user change. The records should connect every action to the relevant person and debt. Your legal and compliance staff should be able to reconstruct the sequence without vendor assistance.

A vendor passes this category only when the software enforces the rules and preserves the evidence. Exportable records allow your legal and compliance staff to retrieve the relevant call, debt, script, and rule history without manually reconstructing events.

Litigation and escalation capability as an evaluation criterion

Escalation capability shows whether a vendor can support unresolved accounts after pre-litigation outreach ends and which work must pass to outside counsel or another provider. Buyers should verify where each vendor’s responsibility ends. Some platforms manage calls, messages, and payment plans, then export unresolved accounts to outside counsel. Others support filing, case administration, court activity, and post-judgment steps within the same platform.

A scorecard should separate escalation coverage from general collection features. Ask vendors to demonstrate how an account moves into small claims, lien enforcement, or other litigation. The demonstration should show which documents the platform prepares, which approvals require human review, how filing deadlines are tracked, and how case records connect with earlier communication logs. Buyers should also identify tasks that still require separate law firms, collection agencies, or manual data entry.

Small claims and lien matters provide a useful economic stress test. Legal fees, court costs, and administrative work can make lower-balance cases uneconomic under a buyer’s cost model. A vendor should quantify which tasks its automation performs and show how those changes affect labor, outside-counsel fees, court costs, timelines, and minimum viable claim values. Buyers should request the assumptions behind any answer, including internal labor, court costs, outside counsel fees, expected timelines, and minimum claim values.

Score vendors on demonstrated coverage rather than a broad claim that they “support litigation.” A higher score should require continuity through filing and court, clear responsibility at each stage, complete audit records, and credible cost modeling for smaller claims. A platform that ends at referral may still fit your operating model, but the scorecard should make the resulting vendor handoffs and costs visible.

Where Delos fits

Buyers evaluating Delos should apply the same compliance-enforcement tests used for every vendor. Ask Delos to demonstrate whether its platform blocks an eighth call within the applicable seven-day period, suppresses calls during the post-conversation period, and restricts limited-content message templates. The Regulation F call-frequency and message requirements define the test criteria, but they do not establish that Delos provides these controls.

Delos describes a model that combines amicable outreach with litigation automation, including small claims and lien-related matters. Buyers should ask Delos to identify the exact stages its product supports, demonstrate how a case moves through filing and court, and disclose where law firms or manual work remain necessary. Any claim that Delos makes smaller claims economically viable requires validation against the buyer’s labor, court-cost, outside-counsel, timing, and recovery assumptions.

If Delos includes its voice onboarding agent in the proposed onboarding or intake process, buyers should request approved product documentation and test how the agent gathers required information, validates it, handles exceptions, and transfers records to the next step. The voice onboarding agent should be scored only on that documented handoff capability. It does not provide evidence of Delos’s recovery performance, pricing, compliance enforcement, or legal outcomes.

Comparison framework: scoring categories at a glance

Use this criteria table to compare vendors against the same operational tests. Score demonstrated system behavior rather than policy statements or sales claims.

Implementation considerations after the scorecard is filled out

Use the scorecard to select vendors for a controlled pilot before approving an enterprise-wide deployment. Start with a limited account segment, named business owners, and defined escalation procedures. Your rollout plan should also specify who approves scripts, rule changes, and exceptions.

Your data migration must preserve the records that drive cadence controls. Map each person, debt, telephone number, call attempt, and completed conversation to the vendor’s data model. Reconcile duplicates and missing timestamps before launch because incomplete histories can cause the platform to calculate rolling call windows incorrectly.

Multi-state configuration requires ongoing compliance ownership. A state-mandated voicemail disclosure may fall outside the content permitted for a limited-content message under Regulation F. Your compliance owner should review state rule changes, update approved templates, test overrides, and record each approval.

Before committing to a full contract, run a pilot on a controlled account segment at real call volume. Test whether the platform blocks an eighth call within seven consecutive days and starts the required cooldown after a telephone conversation. Then export the audit log and confirm that an independent reviewer can reconstruct every attempt, block, conversation event, script version, and rule change. Set acceptance criteria before the pilot so sales demonstrations cannot substitute for measured system behavior.

FAQs

What counts as a Regulation F violation in an automated dialing system?

A collector is presumed to violate Regulation F by placing more than seven calls within seven consecutive calendar days about one debt. The same presumption applies when the collector calls within seven days after a telephone conversation about that debt, according to the CFPB Debt Collection Rule FAQs.

How should enterprise buyers weight compliance enforcement versus recovery performance in an RFP score?

Treat compliance enforcement as a threshold requirement. Compare recovery performance only among vendors that can hard-block prohibited calls, control voicemail content, and produce usable audit records.

Can a single platform handle both early-stage outreach and litigation escalation, or do enterprises need separate vendors?

A single platform can cover both stages, but buyers should verify exactly where the vendor’s service ends. Delos says it combines amicable outreach with litigation automation covering small claims and lien-related matters. Buyers should confirm those capabilities directly during evaluation.

What documentation should a vendor provide to prove call-cadence and voicemail compliance during a pilot?

Require call-level logs that show timestamps, debt identifiers, conversation events, blocked attempts, and timer changes. The vendor should also provide locked voicemail templates, approval histories, and state-specific rule configurations.

Where can buyers find sourced recovery-rate or cost-per-dollar-collected benchmarks to complete this scorecard?

This framework does not cite a specific ACA International or Federal Reserve dataset that provides a directly comparable recovery-rate or cost-per-dollar-collected benchmark. Before using an external figure, verify that its account type, placement stage, measurement period, exclusions, denominator, and outcome definition match the vendor cohort. Require each vendor to provide the same methodology details and its sample size.

Enterprise evaluation decision

A defensible platform decision depends on observed system behavior rather than policy documents or sales claims. Put the scorecard into the RFP, then require each vendor to demonstrate hard compliance controls and export the supporting audit logs during a live demo or pilot.

Score litigation escalation separately because outreach metrics do not reveal which filing, court, post-judgment, or outside-counsel tasks the vendor supports. Ask vendors to demonstrate how a case moves into small claims, lien enforcement, or court, including the work and cost that remain outside the platform. Without that test, a comparison may favor similar outreach features while overlooking a major difference in recovery scope.