Commission Leakage Audits for Broking Firms: Finding Unbilled Endorsements, Missed Renewal Credits, and Insurer Short-Payments

Commission leakage is income the firm earned but never collected: endorsement brokerage that was never billed, renewals that moved without the commission following, statement lines short-paid and never disputed, clawbacks taken twice. In most broking firms it runs quietly at 1 to 4 percent of annual commission income, which was tolerable when yields were fat and is not tolerable now. A firm earning INR 40 crore of commission that recovers even the midpoint of that range adds roughly INR 1 crore of EBITDA with no new revenue, no new headcount, and no client conversation.

The regulatory cycle has made the audit more valuable still. The IRDAI (Payment of Commission) Regulations, 2023 replaced fixed product-wise caps with insurer board-approved commission policies, so rates now vary by insurer and change mid-year as insurers manage the roughly 30 percent expenses-of-management ceiling under the IRDAI (Expenses of Management, including Commission, of Insurers) Regulations, 2024 (35 percent for standalone health insurers). Every rate revision is a leakage opportunity: policies booked at superseded rates, statements paid at rates the broker never verified. And the draft IRDAI (Insurance Intermediaries) (Amendment) Regulations, 2026, which would make intermediation revenue an audited, published schedule, means unexplained gaps between expected and received commission stop being a private inefficiency and start being an audit finding.

A leakage audit is not a reconciliation redo. Reconciliation asks whether the books match the statements. The leakage audit asks a harder question: whether the statements themselves contain everything the firm was entitled to. The method below is designed for a mid-size firm (INR 10 to 75 crore commission income) and runs in six to ten weeks with two to three people, most of them part-time from finance and operations.

Endorsements are the largest leakage family in nearly every audit, because they combine high volume, small individual amounts, and a processing path that often bypasses the commission workflow. A premium-bearing endorsement (sum insured increase, location addition, mid-term cover extension, declaration adjustment on open covers) generates additional premium and therefore additional brokerage. The failure mode is simple: the servicing team processes the endorsement with the insurer, the client pays, and no commission expectation is ever raised because the endorsement was handled over email rather than through the policy administration system.

Where it concentrates:

• Marine open covers and declaration policies, where monthly or quarterly declarations adjust premium continuously and commission should track every adjustment.
• Group health mid-term additions, where employee additions and deletions generate endorsement premium monthly; on a 3,000-life account the monthly endorsement premium can exceed INR 5 lakh, carrying brokerage the firm never sees if the endorsement sits outside the commission table.
• Project and construction policies, where sum insured escalations and period extensions arrive late in the project and are processed under deadline pressure.
• Multi-location property programmes, where location additions are frequent and individually small.

The audit test: pull the insurer's endorsement listing for the audit period (most insurers can produce endorsement-level premium registers for a broker's book on request), match against the firm's commission events, and list every premium-bearing endorsement with no corresponding commission row. In first-time audits, 30 to 60 percent of premium-bearing endorsements on serviced commercial accounts commonly lack a commission expectation. Recovery is usually straightforward because the entitlement is clear; insurers pay against a documented endorsement schedule, though statements older than 24 months meet increasing resistance.

The second family is renewal-linked. A policy the firm placed and serviced renews, but the commission does not arrive, or arrives at the wrong rate, because of how the renewal was coded at the insurer's end.

Four patterns recur. First, broker-code drops: the renewal is issued under the insurer's direct code or another intermediary's code, often after the client dealt with the insurer's branch directly for a routine renewal, and the incumbent broker's servicing continues unpaid. Second, rate resets on renewal: the insurer's commission policy changed between the original placement and renewal, the renewal pays at a lower rate, and nobody at the broker verified the renewal rate against the communicated terms. Under the post-2023 framework, where rates move with insurer board decisions rather than a static regulatory table, this pattern has grown sharply. Third, co-broking splits misapplied, where a shared account renews with the split reversed or omitted. Fourth, long-term policies, such as multi-year fire or home covers attached to commercial relationships, where year-two and year-three commission instalments simply never appear on statements.

The audit test is a renewal survivorship run: take every policy the firm was broker of record on in the prior year, identify which renewed with the same insurer (insurer renewal listings or client confirmation), and check that a renewal commission event exists at the correct current rate. The gap list splits into recoverable items (code errors, rate errors, missing instalments, where the firm's mandate and servicing record support the claim) and commercial losses (genuine broker-of-record changes, which are not leakage but belong in the retention analysis instead).

The third family lives inside the insurer statements the firm already receives. Items appear, but at the wrong amount, and the differences are individually too small to chase, which is exactly why they persist.

The recurring forms: commission calculated on net premium where the agreed basis was gross of certain loadings, or on premium excluding components that carry brokerage; the insurer's system applying a default product rate lower than the negotiated rate for the broker; reward or additional remuneration accrued under an agreed arrangement but never actually credited, particularly where the trigger (volume threshold, persistency level) was met late in the year; GST on brokerage mishandled between the parties so the credited amount is 18 percent adrift of expectation; TDS under Section 194D deducted at a rate inconsistent with the certificate position for the period; and double clawbacks, where a cancellation reversal is taken on two consecutive statements.

Short-payments are a percentage game. Individually they run from a few hundred rupees to a few thousand; across a year of statements from 15 to 25 insurers they compound. The audit test is a full-year re-computation: for a sample (or, with decent data, the full population) of statement lines, recompute expected commission from the rate master and premium records, and age every negative variance beyond tolerance. Sort the output by insurer and by cause, not by policy, because recovery happens insurer by insurer: a single consolidated variance schedule presented to an insurer's broker-servicing team recovers far more, at far lower cost, than item-by-item emails from account handlers.

Variance can also run in the firm's favour through overpayments and duplicate credits. Return them proactively: a firm chasing INR 40 lakh of short-payments while sitting on known overpayments has donated its negotiating position, and unreturned excess credits are a fraud-control finding waiting to happen.

The method assembles the three family tests into a single engagement. For a mid-size firm the sequence is:

1. Scope and data assembly (week 1-2). Fix the audit window, normally the trailing 24 months, since older claims recover poorly. Assemble the commission events data, the general ledger, all insurer statements, the rate master with its change history, the placement and renewal registers, and endorsement listings requested from the top 10 insurers by volume, which will typically cover 80 percent or more of income.
2. Completeness pass (week 2-4). The endorsement test and the renewal survivorship test: entitlement that never became an expectation. This pass finds the largest rupee amounts.
3. Accuracy pass (week 3-5). Re-computation of statement lines against the rate master: expectation that was underpaid. Run reward and additional-remuneration arrangements as a separate worksheet, checking each trigger against actual volumes.
4. Collection pass (week 4-6). Ageing of confirmed-but-unpaid items already in the books; commission receivables beyond 90 days are leakage in slow motion and belong in the same recovery pack.
5. Recovery execution (week 5-10). One consolidated schedule per insurer, with supporting evidence attached item by item: endorsement copies, renewal schedules, rate communications, mandate letters. Escalate through the insurer's broker-relationship channel first and its finance channel second. Agree settlements in writing and post them against the original events, not as miscellaneous income, so the audit trail closes.
6. Root-cause fixes (week 8-10). Every recovered rupee has a process defect behind it. The standard fixes: endorsements routed through the policy administration system without exception, a monthly renewal survivorship report, rate-master updates within five working days of any insurer communication, and statement re-computation built into the monthly close.

Staffing reality: one finance analyst near full-time, one operations person half-time, and four to six hours a week of a senior owner for insurer escalations. Firms with clean policy-number capture finish nearer six weeks; firms matching on client names finish nearer ten.

Recovery expectations keep the engagement honest, so set them from observed ranges rather than hope. For a firm with INR 25 crore of annual commission income running its first structured leakage audit over a 24-month window, indicative outcomes based on broker operational patterns look like this:

• Unbilled endorsement commission: typically 0.5 to 1.5 percent of annual commission income identified, of which 70 to 85 percent is recoverable because entitlement is documentary. On INR 25 crore, roughly INR 12 to 38 lakh identified, INR 10 to 30 lakh recovered.
• Renewal credits and code errors: 0.3 to 1.0 percent identified; recovery near 80 percent inside 12 months of the renewal, falling steeply for older items. Roughly INR 8 to 25 lakh identified, INR 5 to 18 lakh recovered.
• Short-payments and statement mismatches: 0.2 to 0.8 percent identified; recovery 60 to 80 percent, lower where the dispute is about the premium basis rather than an arithmetic error. Roughly INR 5 to 20 lakh identified, INR 3 to 14 lakh recovered.
• Reward arrangements never credited: lumpy and firm-specific; zero in some firms, north of INR 25 lakh in firms with several volume arrangements and weak tracking.

A realistic total for a first audit: identification of 1 to 3 percent of annual commission income and cash recovery of 0.7 to 2 percent, so INR 18 to 50 lakh on the INR 25 crore firm, against an internal cost of a few lakh in staff time. Second-year audits find half as much, which is the point: the value shifts from recovery to the monthly controls that stop the leak.

The audit is a one-time cash event; the durable value is the control set it leaves behind. Four controls, each cheap, close the large majority of future leakage:

Endorsement discipline. No endorsement is requested from an insurer except through the policy administration system, which auto-creates the commission expectation. Email-processed endorsements are the root cause of the biggest leakage family, and the fix is a workflow rule plus a monthly exception report of insurer-confirmed endorsements with no system record.

Renewal survivorship reporting. A monthly report of last month's renewals due versus renewed-and-commissioned, with broker-code and rate verification on each. This doubles as retention MIS and catches code drops inside the correction window.

Rate-master governance. A single owned table of insurer commission terms with effective dates and the source communication attached, updated within five working days of any change. Under the 2023 framework, where insurer boards revise terms as EOM pressure dictates, a stale rate master silently converts rate cuts into unexplained variance and rate rises into leakage.

Quarterly mini-audits. A one-week version of the full method each quarter: endorsement match against the top five insurers, survivorship on the quarter's renewals, and re-computation of one month's statements. Firms running the quarterly rhythm report steady-state leakage below 0.5 percent of commission income.

There is a governance dividend too: the same evidence pack that recovers money is the substance of audit readiness for the proposed intermediary disclosure regime. Leakage work is the rare project that pays for itself in cash while building the compliance asset the next two years of regulation will demand.