Multiple beneficiary rule

A multiple beneficiary rule detects one account sending money to an unusually large number of distinct beneficiaries inside a time window. It counts the breadth of destinations rather than the volume of payments, which separates an account fanning money out from one repeatedly paying the same few payees.

The fraud scenario this rule addresses

Fraud proceeds are often split across many accounts to make each movement look small and to reduce the loss if one receiving account is frozen. An account that receives a lump of stolen money and distributes it to many different recipients, each getting a modest amount, is performing the fan-out step of layering.

A taken-over account can show the same pattern when an attacker sprays funds to several accounts they control to avoid per-payee limits.

Data the rule requires

Example logic

Illustrative pseudocode: it compares each event with the rule's parameters and flags the ones that match. Parameters are symbolic: every value is institution-specific and has to be derived from your own transaction distribution.

RULE multiple_beneficiary
  FOR EACH payment P from account A:
    window_payments = payments WHERE sender = A
                                 AND timestamp > P.timestamp - :window
    distinct_payees = count_distinct(window_payments.beneficiary)
    IF distinct_payees >= :max_distinct_beneficiaries
       AND sum(window_payments.amount) >= :min_total:
      RAISE alert(rule = "multiple_beneficiary",
                  account = A, evidence = window_payments)

Choosing a threshold

The count of distinct beneficiaries and the window are the two parameters, with a minimum total that stops trivial activity qualifying. A short window with a low count fires on a customer settling a restaurant bill between friends. A long window with a high count lets an attacker spread the fan-out over days.

Beneficiary identity matters more than it appears. If the same real payee appears under different account references, the count is inflated, and if different people share an identifier, it is deflated. How payees are normalised therefore changes the effective threshold.

Setting the limit from the account's own history works better than a global figure, because a payroll-style account legitimately has a very large payee count while a retail account normally has a handful.

False positives to expect

Payroll and benefit runs, a business paying many suppliers, a charity distributing grants, a marketplace paying out its sellers, a customer splitting a group dinner or holiday bill among friends, and a family sending remittances to many relatives all produce many distinct beneficiaries legitimately.

These are largely business and payment-service accounts, so excluding or separately profiling accounts whose normal behaviour is high-fan-out removes most of the noise. Consider also the value distributed: genuine fan-out tends to match known payee lists the account has used before.

Investigating an alert from this rule

  1. List the beneficiaries and check how many are new to the account compared with established payees.
  2. Check the source of the money being distributed: did a large credit arrive shortly before the fan-out?
  3. Compare the amounts. Equal or near-equal amounts to many payees suggest a systematic distribution rather than ordinary spending.
  4. Check whether the beneficiaries share characteristics, such as being newly opened or held at the same institution.
  5. Check whether the account type or customer profile explains the behaviour, such as a payroll or settlement function.
  6. Review the beneficiaries' own activity for signs they are part of the same network, and tag them if so.

Related rules

Where this rule fits

The use case this rule supports is covered in mule accounts, and the fraud rules engine page describes how conditions like these are expressed and versioned.

Common questions

What is a multiple beneficiary rule?

A multiple beneficiary rule is a fraud rule that raises an alert when one account sends money to an unusually large number of different payees inside a time window. It counts distinct destinations rather than payments. That fan-out is how fraud proceeds are split across many accounts to keep each movement small and limit the loss if one is frozen.

The count depends on how payees are identified. If one payee appears under several account references the count is inflated, so identifiers have to be normalised before the threshold means anything.

Payroll, supplier payments and marketplace payouts legitimately reach many payees, so the rule is usually set against the account's own history, or applied only to account types that do not normally behave that way.

Discuss your transaction monitoring requirements

Tell us which transaction flows you need to monitor, which systems Akoben would integrate with, and how your fraud team works today.

Request a demo