---
title: "Collections software quotes: count active customers before comparing offers"
canonical: https://www.billabex.com/en/blog/collections-software-quote-active-customers/
lang: en
alternate: https://www.billabex.com/fr/blog/devis-logiciel-relance-clients-actifs.md
alternates:
  fr: https://www.billabex.com/fr/blog/devis-logiciel-relance-clients-actifs.md
  en: https://www.billabex.com/en/blog/collections-software-quote-active-customers.md
  es: https://www.billabex.com/es/blog/presupuesto-software-recordatorios-contar-clientes-activos-antes-de-comparar.md
updated: 2026-11-06
index: https://www.billabex.com/llms.txt
---

# Collections software quotes: count active customers before comparing offers

A supplier asks how many customers you manage. You answer four hundred, using the row count in an export. Those rows actually represent invoices, several of which belong to the same businesses. Only some identified customers will receive a reminder during the month. The quotation can be internally consistent with your answer while still being poorly sized for the intended use.

Before comparing collection software offers, prepare a measure that matches each supplier's pricing unit. An active customer can mean different things. Establish the event that creates activity, the period being measured and how customer identities are grouped. This prevents a lengthy price discussion based on the wrong quantity, and makes the assumptions behind each quotation reviewable by the people who will operate the system.

## Start with the contractual definition

Billabex's terms define an active debtor through receipt of at least **one email sent by an AI agent during the billing period**. Article 9 refers the price and applicable scale to the quotation or signed specific terms. Imported invoice count alone is therefore not this unit. [Source: Billabex terms, Articles 4 and 9, version dated 17 June 2026](https://www.billabex.com/en/terms-and-conditions-of-use/).

For any offer, clarify edge cases: an imported but uncontacted customer, a rejected email, multiple recipients at one company, historical data and a change of billing period. Ask how the count can be explained from underlying events. Do not silently replace a contractual receipt-based definition with an assumption about pressing a send button or creating a draft.

An active metric in another application cannot automatically be reused. Stripe Billing, for example, allows subscriber activity to begin at subscription start or when the first payment arrives. That concerns its subscription analytics, rather than the price of collection software. [Source: Stripe Billing documentation](https://docs.stripe.com/billing/subscriptions/analytics).

## Reconcile identities before measuring activity

A useful export should identify the debtor business, its source-system reference and the relevant events. Multiple invoices, several contacts or variations in a company name can produce several rows for one debtor. Conversely, a commercial group may contain separate companies that should not be merged merely because their names or contacts look related.

The distinction between [legal entities within a customer group](https://www.billabex.com/en/blog/payer-versus-customer-entity-mapping/) helps structure this reconciliation. Use available identifiers and document mappings between applications. A shared email address does not establish that two records represent one customer. A changed address should not automatically create a new debtor either. Unresolved identity cases should remain visible in the assumptions, rather than disappearing into a rounded estimate.

The UK Government Data Quality Framework distinguishes **six dimensions**, including uniqueness and accuracy. It explains that complete records can still contain incorrect information. We apply that distinction to preparing a count; the framework does not define any supplier's commercial pricing units. [Source: Government Data Quality Framework](https://www.gov.uk/government/publications/the-government-data-quality-framework/the-government-data-quality-framework).

## Worked example: four hundred rows, sixty active customers

Consider a fictional portfolio containing **400 invoice rows**, reconciled to **150 distinct debtors**. During the monthly period under review, sixty debtors each receive at least one email meeting the definition adopted for the example. Thirty of those sixty receive a second email, and no other emails are included in this simplified scenario.

The result is **90 emails and 60 active debtors**, not four hundred active customers. Four hundred rows describe document volume. One hundred and fifty debtors describe the portfolio. Sixty active debtors describe activity during the period. A supplier may need all these figures, but they should not be presented as interchangeable measures of the same thing.

Portfolio activation is **60 / 150 = 40%** in this example. That is neither an industry average nor Billabex performance. It follows solely from the fictional month's assumptions. Using four hundred invoices as the customer count could overstate the customer volume. Using an unusually quiet month without explaining it could understate future requirements, even if the count itself is accurate.

Keep enough detail to reproduce the calculation: debtor identifier, period and qualifying event. An initial commercial estimate does not necessarily require sharing complete customer conversations. Aggregated volumes and a clear explanation of the counting rules can frame the discussion, with a more detailed data exchange only if the subsequent assessment actually requires it.

## Do not add unique customers like invoice amounts

Suppose the next month has **75 active debtors**, of whom **35** also appeared among the first month's sixty. Across both months, the number of distinct debtors is **60 + 75 − 35 = 100**. Adding the monthly observations produces **135**, but that does not represent one hundred and thirty-five different businesses.

For a monthly unit, report **60 followed by 75**. For distinct customers across the two-month period, report **100**. The observed monthly maximum is **75**, while the monthly average is **67.5**. An average can contain a decimal without anybody billing half a customer; it is a planning measure, not a contractual rounding rule or an assumed invoice calculation.

Ask what happens when activity crosses a tier and when it subsequently falls. The answer may involve commitments, plan changes, minimums or overage treatment. Do not assume the same mechanism across suppliers. Equally, an annual subscription term does not automatically mean that unique debtors should be accumulated across a full year for pricing purposes.

## Show seasonality and the initial backlog

Representative history should include the periods that explain your business: billing campaigns, financial closes, customer holidays and renewals. Present individual months with that context. An annual average may conceal a peak that matters for operational capacity and support, even where the commercial tariff uses another measure. Make that peak visible without presenting it as normal every month.

Starting the project may also concentrate an older backlog. Separate historical accounts to be examined, customers that can actually be contacted and the expected ongoing flow. A [replayed import without duplicate reminders](https://www.billabex.com/en/blog/replayed-import-duplicate-reminders/) should preserve identities. It should not inflate your planning estimate by recreating the same customers each time data is loaded or an interrupted import is resumed.

The scenario should reflect the intended collections strategy. Do not assume every imported customer will be contacted immediately: some balances are paid, disputed or insufficiently documented. Conversely, do not quietly exclude difficult accounts just to produce a smaller number. Explain exclusions and their reasons so the quotation remains comparable with the project that staff will actually run.

## Compare offers against the same scenario

Give each supplier the same base scenario, then request its count under its own rule. Results can legitimately differ. Explain that difference before comparing totals: users, debtors, invoicing entities, integrations and implementation support may be separate pricing components. A lower headline number is difficult to interpret if it covers fewer activities or relies on a different definition of use.

Add a peak scenario and a lower-activity scenario. Request an explicit amount, its scope and the conditions that would change it. [Useful customer replies](https://www.billabex.com/en/blog/useful-reply-rate-reminders/) help assess the work an application supports; they should not be confused with the unit triggering its charges. Operational results and commercial measurement both matter, but they answer different questions.

To prepare a [debt collection software](https://www.billabex.com/en/debt-collection-software/) quotation, bring monthly volumes, the definitions used and unresolved cases. The supplier should be able to connect its offer to those assumptions. That gives your team a reproducible calculation for the intended usage and a practical basis for checking future invoices, rather than two superficially comparable numbers that merely share the word active.

## Sources

- Billabex, terms dated 17 June 2026: [definitions and financial conditions, Articles 4 and 9](https://www.billabex.com/en/terms-and-conditions-of-use/).
- Stripe: [Billing Analytics and configurable active-subscriber definitions](https://docs.stripe.com/billing/subscriptions/analytics).
- UK Government: [Government Data Quality Framework](https://www.gov.uk/government/publications/the-government-data-quality-framework/the-government-data-quality-framework).