Persistency Management for Individual Advisors: Working the 13th and 25th Month

Most individual advisors measure themselves by what they sold. The number that determines whether ten years of selling produces a business or a treadmill is persistency: the share of what you sold that is still in force, still paying, and still yours.

Two advisors write the same volume for five years. The first keeps 85 percent of it alive past the second year; the second keeps 55 percent. By year five the first has a book that pays them for work done in year one and households they can go back to. The second has a treadmill, because half of everything they wrote must be replaced before it earns a second rupee, and the households they burned are not available to sell to again.

Persistency is also the metric your principal watches most closely and rarely discusses with you. Insurers report it publicly, are measured on it, and build it into how they treat distributors. An advisor who does not know their own 13th-month number is being assessed on a figure they cannot see.

Policy direction is about to make it louder. IRDAI's stated rationale for reworking how commission is paid across a policy's tenure is precisely to incentivise long-term servicing and improve persistency. That reform is a proposal, not a rule, and this post treats it as such. But the metric it targets is one you can protect today at zero cost.

Persistency is measured at fixed intervals from the month a policy commenced, and Indian life insurers report it publicly at the 13th, 25th, 37th, 49th and 61st month. The first two carry most of the weight.

13th-month persistency answers one question: of the policies that started 13 months ago, how many are still in force? A policy sold in June 2025 on annual mode has one renewal due in June 2026, and counts in the 13th-month cohort measured in July 2026. If that premium went unpaid and the grace period expired, the cohort loses it.

25th-month persistency asks the same question a year further out. It captures the second renewal, a much harder test: by month 25 the sale is a memory, the advisor has moved on to newer clients, and nothing about the policy is novel to the household.

Two mechanics get misread.

The basis changes the number. Persistency on policy count treats a INR 5,000 policy and a INR 5 lakh policy identically. On premium, it weights by rupees. An advisor who loses many small policies but keeps the large ones looks fine on premium and poor on count. When your principal quotes a figure, ask which basis it is on before agreeing or arguing.

Premium mode changes the exposure. An annual-mode policy has one payment event before the 13th month. A monthly-mode policy has twelve. Every payment event can fail, and monthly modes fail more often, usually not through decision but through a bounced auto-debit nobody chased.

Run the numbers on one advisor to see why this compounds rather than adds. Take an advisor writing 120 policies a year at an average annual premium of INR 18,000, roughly INR 21.6 lakh of new annualised premium. Compare two persistency profiles across five years, holding sales flat.

At 85 percent 13th-month and 78 percent 25th-month, each year's writing survives into the next, so year five's book is not 120 policies, it is closer to 500, and the renewal premium dwarfs the new business. Year five is largely servicing an existing base, with new sales as growth rather than replacement.

At 60 percent 13th-month and 45 percent 25th-month, the same 120 policies a year produce a book that stalls near 250 and stops growing, because attrition eats each new cohort at roughly the rate the advisor adds one. Year five looks like year three. New sales are not growth, they are patching.

The gap is not talent or luck. It is about forty hours of work a year, in the right two windows.

The hidden costs make it worse. A first-year lapse typically triggers a recovery of remuneration under your engagement contract, so you do not just fail to earn the renewal, you give back part of what you already earned. A lapsed household is a closed door: the client who let a policy die because nobody called is not a warm lead for anything else.

This is the highest-return window in an advisor's year, and almost nobody works it, because the policy is not due yet and nothing is on fire. What kills a first renewal is rarely a decision to cancel. It is four things, all visible and fixable months in advance:

1. The payment fails silently. The mandate sits on an account the client no longer uses, the card expired, the balance was short, the auto-debit was never registered. The client does not know the policy lapsed.
2. They do not remember what they bought. They recall a number and your face, but cannot say what the policy does, so the renewal notice reads as a bill for nothing.
3. The sale was stretched. A premium ambitious against the household's real cash flow survives one payment, not two. This is set at the point of sale and cannot be repaired in month 11.
4. Contact decayed. The number changed, the household moved, the notice went to an old email.

The sequence that fixes all four:

• Month 9: verify the plumbing. Confirm the mandate is live against a current account, the card has not expired, and the registered mobile and email are the ones the household uses. The cheapest persistency work there is, and it fixes the most common cause.
• Month 10: re-establish what they own. One message, in the language the household speaks, saying what the policy covers, pays and excludes. Send the insurer's approved material, not your own words.
• Month 11: name the date and the number. Give them the renewal date and exact premium, and ask one question: is this comfortable this year. If not, you have a month to work with the insurer on mode or options rather than a lapse to explain.
• Month 12: confirm receipt. Not "the renewal is due" but "has it gone through." A policy in grace is still saveable; one past it is a revival case with paperwork.

None of this is complicated. It is a calendar problem: renewal dates live scattered across notebooks, chat threads and whichever insurer portals the advisor still has passwords for, so the month 9 cohort is never visible on any given morning.

Advisors who fix the 13th month often assume the 25th follows. It does not, because the failure mode changes. By month 21 the payment plumbing has worked once, which removes the mechanical cause. What is left is the relationship, decaying for two years. The second renewal is where the household asks what it did not ask the first time: is this still worth it?

That question gets answered by whatever happened in the two years since the sale. If the advisor called only to sell, the answer trends towards no. If the only contact was the renewal reminder, the policy is competing for cash in a household budget with no advocate.

The 21 to 24 window is about relevance, not payment:

• Check whether the cover still fits. Two years is long enough for a birth, a job change, a loan, a parent moving in. A policy sized against a 2024 household and never revisited is one the client is right to question, and revisiting it is the honest route to a second sale.
• Show the policy doing something. Most policies are invisible until a claim. If the household has claimed on anything you placed, that is the most powerful persistency event available. If not, tell them what would happen if they did, with their numbers.
• Talk to the household, not the payer. A spouse who has never heard of the policy will call it a waste when money gets tight. Advisors with strong 25th-month numbers are known to the whole household, not one contact.
• Do not replace your own policy. Funding a new sale by lapsing an existing one destroys your persistency, triggers a recovery on the old contract, and is precisely the churn the regulator says it wants to stop.

There is a reason to do this now beyond the arithmetic, and it needs stating precisely, because there is a lot of loose talk about it.

In force today: the IRDAI (Payment of Commission) Regulations, 2023 removed product-wise caps from April 2023, leaving commission to each insurer's own policy within the ceilings set by the IRDAI (Expenses of Management, including Commission, of Insurers) Regulations, 2024, effective 1 April 2024. The Sabka Bima Sabki Raksha (Amendment of Insurance Laws) Act, 2025, in force from 5 February 2026, restored IRDAI's power to cap distributor commissions and made intermediary licences perpetual. The Act sets no cap, and none has been made under it.

Not decided: as of this post's date, IRDAI's commission consultation paper has not been published. Chairperson Ajay Seth indicated in early July 2026 that it is expected by end-July. Business Standard (3 July 2026) and Business Today (9 July 2026) describe ideas said to be under consideration: spreading commission across the policy life instead of concentrating it upfront, differentiating remuneration by the depth of service an advisor gives, product-wise caps varying by complexity and tenure, and tighter disclosure. Every one is a proposal reported in the press. None is a rule.

On the record is the regulator's reasoning: spreading commission across a policy's tenure is meant to incentivise long-term servicing, improve persistency and strengthen trust. The concern is that paying a large share at the point of sale pushes volume over suitability. July 2026 reporting notes distributors can currently earn up to roughly 40 percent of premium on some life and health products, substantially upfront: an observed market level, not a regulatory cap.

The asymmetry makes this easy. If nothing changes, high persistency still pays on today's rules. If something like the reported direction arrives, an advisor already servicing their in-force book is positioned for it and one living on first-year concentration is not. There is no version of the future in which working months 9 to 12 was the wrong call, and it needs no paper published first.

You cannot work a metric you cannot see, and the reason most advisors do not manage persistency is not indifference. The data is scattered across places that do not talk to each other: an insurer portal for one book, a WhatsApp thread for another, a notebook for the rest, memory for the awkward cases.

The minimum measurement is one list with six fields per policy: client, insurer, policy number, commencement month, mode, and premium. Commencement month gives you the 13th and 25th month automatically. Mode tells you how many payment events sit before each. Premium lets you compute both bases and see whether you are losing small policies or large ones.

With that list, four views run the year:

1. The month-9 cohort. Every policy that commenced nine months ago. Your plumbing-check list.
2. The month-21 cohort. Every policy approaching its second renewal. Your relevance-check list.
3. In grace, right now. Worked weekly, not monthly.
4. Lapsed in the last ninety days, with cause tagged. The list that tells you what to fix.

That fourth view earns its place. Every failure has a cause (mandate failed, client moved, income dropped, spouse objected, bought elsewhere, never understood the product) and it is almost always recorded nowhere. Tag each lapse in one word as you find out, and after a year you have a distribution rather than a feeling: 70 percent failed mandates is a plumbing problem that belongs in month 9, 70 percent "stopped seeing the value" is a relevance problem that belongs in month 21. Without tags, advisors guess in one direction. They assume clients chose to leave, because a silent auto-debit failure generates no conversation and no memory.

Then compute your own 13th and 25th month figures each quarter, on both bases, and hold them next to your sales figure. Most advisors have never seen these two numbers side by side, and the moment they do, the allocation of their week changes without anyone arguing for it.

An advisor who walks in carrying a documented 13th-month figure, a cause distribution for their lapses and a servicing calendar is negotiating from a different position than one carrying a sales number. Licences are perpetual now, so the advisor-principal relationship is commercial and gets renegotiated on evidence. Persistency is the evidence, and it is the same work that makes the book worth having.