Building the Servicing Effort Evidence Base Before Effort-Based Commission Arrives

In early July 2026, IRDAI signalled that it is preparing an overhaul of commission rules aimed at curbing mis-selling, with a consultation paper expected around end July per remarks attributed to Chairperson Ajay Seth. Among the ideas under discussion: staggered or trail commissions over the policy life, and effort-based remuneration paying more for advisory, documentation, and claims servicing than for passive distribution channels. A companion post in this series analyses the consultation itself; this one is about the other half of the preparation, the operational data build.

The uncomfortable premise: every broker believes it earns its commission through servicing effort, and almost none can prove it. Ask a mid-market firm to produce, for a named client, last year's endorsement count with turnaround times, the claims it advocated with outcomes and hours, the renewal review minutes, and the advisory notes issued, and the honest answer at most firms is a mailbox archaeology project. The work happened; the record did not.

An effort-based regime, whatever final form it takes, cannot pay for effort directly. It will pay for evidenced effort: records that a remuneration committee, an insurer's audit team, or an IRDAI inspector can verify. The same evidence already has three uses today, before any rule changes: it supports higher servicing-tier rates inside insurers' board-approved commission grids under the IRDAI (Payment of Commission) Regulations, 2023, it defends the firm's revenue recognition judgements where commission is conditioned on servicing, and it wins renewals by making the stewardship case concrete. A firm that starts capturing in Q2 of FY 2026-27 walks into any new regime with a year of evidence; a firm that waits for the final regulation starts from zero on the day the regime begins.

The consultation paper will define categories in its own terms, but the direction signalled (advisory, documentation, claims servicing valued above passive placement) maps onto five working categories a broker can structure its capture around now.

1. Advisory effort: risk reviews, programme design work, sum insured and business interruption adequacy reviews, market and regulatory updates issued to clients, coverage gap analyses, comparisons prepared across quotes with recommendations and reasons.
2. Documentation effort: proposal preparation, policy checking against quoted terms (with errors caught and corrected), wording negotiations, certificate and endorsement issuance, KYC and compliance documentation handled.
3. Claims servicing effort: intimation handling, surveyor coordination, document collection and follow-up, settlement negotiation support, dispute escalation, claims MIS to clients.
4. Renewal and lifecycle effort: renewal strategy reviews, remarketing exercises with markets approached and outcomes, mid-term programme adjustments, endorsement handling volumes and turnaround.
5. Relationship and governance effort: stewardship meetings held, escalations resolved, service level reporting delivered.

Two design principles matter more than the taxonomy itself. First, capture must be per client and per policy, because remuneration attaches to placements: firm-level activity totals prove nothing about any specific commission. Second, capture must record outcomes alongside activity: an endorsement processed in 2 days versus 15, a claim settled at 94 percent of assessed loss versus abandoned, a policy-checking error caught before issuance. Activity without outcomes reads as motion; outcomes are what distinguish advisory value from passive channel behaviour, which is precisely the distinction the proposed regime exists to draw.

Most broker CRMs in India are deployed as pipeline tools: prospects, quotes, wins. Converting the CRM into a servicing evidence system is mostly configuration and management discipline, not new software.

The minimum capture standard

Every substantive client interaction gets a logged activity with six fields: client and policy reference, date and staff member, activity category (from the firm's effort taxonomy), a one-to-three line description of what was done, time spent in bands (under 30 minutes, 30 to 120, over 120), and outcome or next action. The bands matter: precise timesheets fail culturally at broking firms within weeks, while banded effort capture survives because it takes ten seconds.

Making it stick

1. Log at the point of work, not at week-end. Retrospective logging collapses into fiction. Integrate capture into the tools where work happens: email plug-ins that file client correspondence against the policy record, telephony logs linked to client records, meeting notes templated in the CRM.
2. Make the account manager's monthly review depend on it. If the servicing dashboard reviewed in monthly operations meetings is built from CRM activity data, the data gets entered. Evidence capture enforced only by policy dies; enforced by the firm's own management cadence, it survives.
3. Sample-audit quarterly. An operations lead pulls 20 random activities and checks them against the underlying emails and documents, publishing the accuracy rate internally. The audit converts a CRM log from a self-serving diary into something the firm can hand to an insurer or regulator with a straight face.

A realistic benchmark for a mid-market firm running this discipline: 60 to 80 logged activities per corporate client per year on a serviced commercial account, versus the 5 to 10 a pipeline-configured CRM typically holds. That gap is the difference between asserting effort and demonstrating it.

Claims servicing is the effort category regulators care about most, because it is where policyholder harm from passive distribution shows up. It is also where broker evidence is weakest: claims work happens under time pressure, across phone calls and site visits, and the file that results is usually the insurer's file, not the broker's.

A broker claims-advocacy record, kept per claim, should contain:

1. Timeline entries: intimation date and channel, date the broker notified the insurer, surveyor appointment date, each document requested and the date the broker obtained and submitted it, each follow-up made, settlement offer dates and amounts, closure date.
2. Intervention notes: where the broker changed the outcome. The surveyor's initial assessment challenged with policy wording arguments, the underinsurance computation corrected, the repudiation contested with precedent, the interim payment negotiated during business interruption. These notes are the advocacy evidence; a timeline alone shows attendance, not advocacy.
3. Outcome metrics: claimed amount, assessed amount, settled amount, days from intimation to settlement, and the same measures for the client's prior unadvocated claims where they exist.
4. Effort measures: visits made, hearings or meetings attended, hours in bands.

Aggregated, these records produce the statistics that will matter in an effort-based conversation: average settlement ratio on advocated claims, median settlement days versus market norms, count of repudiations reversed or reduced. A firm that can show it handled 140 claims last year with a 91 percent average settlement-to-assessed ratio and a 47-day median cycle is holding a different negotiating position, with insurers today and with the regulator's framework tomorrow, than a firm that says it works hard on claims.

Two further record types complete the core evidence base, and both are cheap because the underlying work already happens.

Renewal review minutes

Every commercial renewal above a threshold (say INR 5 lakh premium) should generate a short structured minute: the date of the review meeting or call, who attended, the expiring programme summary, changes in the client's risk profile discussed, sum insured and coverage adequacy assessment, markets approached with quotes received, the recommendation made and its reasons, and the client's decision. One page, templated, filed against the policy.

The minute proves the advisory category almost by itself. It documents that the renewal was a reviewed decision rather than a rolled-over lapse of attention, which is the exact behaviour an anti-mis-selling remuneration regime is designed to reward. It also happens to be the strongest single defence in a professional indemnity claim alleging the broker failed to advise on adequacy, and the natural anchor for the annual stewardship report sophisticated corporates already expect.

Endorsement handling statistics

Endorsement work is high-volume, invisible, and almost perfectly capturable from systems the firm already runs. For each endorsement: request date and channel, type (addition, deletion, correction, sum insured change), date submitted to insurer, insurer turnaround, date delivered to client, and errors caught in checking. From this, quarterly statistics per client and per insurer fall out automatically: volumes handled, median end-to-end turnaround, broker-side versus insurer-side delay split, error catch rate.

The insurer-side delay split deserves emphasis. When effort-based remuneration arrives, insurers will scrutinise broker service claims; a broker holding data that shows its own median endorsement handling at 1.8 days against a particular insurer's 9-day issuance lag is equipped for that conversation in both directions: evidencing its effort and negotiating the insurer's service standards.

Evidence that will be shown to insurers, auditors, or IRDAI must satisfy tests that internal management data never faces. Five architectural requirements follow.

1. Contemporaneous timestamps. Records must be created at or near the time of the work, with system-generated creation dates. A CRM that allows backdated activity entries without an audit trail produces evidence a reviewer can discount wholesale. Most platforms can restrict or log backdating; switch it on.
2. Attribution. Every record tied to a named staff member through individual logins. Shared team accounts destroy attribution and, with it, credibility.
3. Policy-level linkage. Every activity, claim record, minute, and endorsement statistic keyed to client and policy identifiers consistent with the broking system, so effort can be joined to the commission it justifies. This join is the whole game: remuneration frameworks will operate at product and policy level, and evidence that cannot be joined to a policy cannot support its commission.
4. Immutability of closed records. Claims records and renewal minutes should be locked or version-controlled on closure. An editable history is a challengeable history.
5. Retention and export. Keep records for at least the broking regulation record-keeping horizon, and make sure they export in structured form. If the consultation hardens into rules requiring effort reporting, the firms that can produce a clean extract per insurer per period will comply in days.

None of this requires new platforms at most firms. The typical mid-market stack (a CRM, a broking administration system, an email server, spreadsheets around claims) already holds fragments of everything described. The project is consolidation and discipline: one taxonomy, mandatory fields, linked identifiers, locked histories. A realistic build for a 50-to-150 person firm is one quarter of configuration and training, at a cost dominated by management attention rather than licence fees.

The full evidence base described above is a year's journey, but the first 90 days determine whether it happens at all.

1. Days 1 to 15: taxonomy and templates. Adopt the five effort categories, define the six-field activity standard, and issue the claims record and renewal minute templates. Decide the thresholds (which accounts, which claims, which renewals) so the discipline lands first where the commission is concentrated: the top 50 clients by brokerage typically cover 70 to 80 percent of revenue.
2. Days 16 to 45: system configuration. Configure CRM activity types, mandatory fields, and policy-reference linkage; restrict backdating; build the endorsement statistics extract from the administration system; stand up the claims record library.
3. Days 46 to 90: pilot on the top accounts. Run full capture on the top-client segment, review weekly, fix the friction points (the fields nobody fills, the workflow steps that get skipped), and publish the first monthly servicing dashboard to leadership.

After the pilot, governance carries it. Three mechanisms are enough: the monthly operations review built on the servicing dashboard so the data has an internal customer, the quarterly sample audit so the data stays honest, and an annual evidence pack per major insurer relationship (activity summaries, claims statistics, endorsement turnaround) used in grid negotiations so the data earns revenue.

The strategic point deserves stating plainly. Effort-based remuneration is a proposal; the consultation paper may land differently, and final rules will take time. But every use of this evidence base pays off under current rules: servicing-tier commission cases inside board-approved grids, revenue recognition support, renewal defence, professional indemnity protection, and stewardship reporting. The data build is a no-regret move with a regulatory option attached. Brokers who wait to see the final regulation will be assembling in a hurry, from mailboxes, the record their competitors have been compounding for two years.