A branch manager in Pune told me his best agent had the worst dashboard. She closed ₹14 lakh of premium last quarter across motor and health. Her call count was the lowest on the floor. The agent above her on the leaderboard made 312 calls, closed ₹4 lakh, and got the appreciation email. That is what broken insurance agent performance tracking does — it rewards the dialler and quietly demoralises the closer.
Most insurance branches still measure agents on activity: calls made, leads touched, hours logged. It is easy to count and almost useless. This is about measuring agents on what they actually recover, how fast they reach the leads worth reaching, and how to run a 1:1 that the agent can't argue with because the data is sitting in front of both of you.
TL;DR
- Call counts reward effort and punish results. An agent who burns 80 dead leads beats one who closed three ₹1.5L policies in four calls. That is backwards.
- Measure three things instead: ₹ premium recovered per agent, response time on Grade A/B leads, and follow-up plus renewal completion.
- A high-grade lead worth ₹2L premium that waits 5 hours usually buys elsewhere. Speed on the right leads is the metric, not speed on all leads.
- A timestamped, per-agent audit trail (calls + WhatsApp) makes 1:1s fair and keeps you compliance-ready without reconstructing anything at quarter-end.
- Leadkaun's Sales Rep Tracking does this per agent across 10–40 reps. ₹999–₹2,999 per rep/month.
Why call counts measure the wrong thing
A call count tells you an agent was busy. It tells you nothing about whether the branch made money. Those are different questions, and conflating them costs premium every month.
Picture a Tuesday. Two health-insurance leads land in the queue at 11:14 AM. One is a 34-year-old salaried buyer comparing a ₹25k family floater — warm, ready, Grade A. The other is a tyre-kicker who filled a form for a quote six weeks ago, never answered since, Grade E. An agent chasing a 60-call daily target will often work the easy dials first. The Grade E lead is "a call" the same as the Grade A. So the floater buyer sits. By the time the agent gets to her at 3:40 PM, she has already taken a call from a competitor and bought. The agent still logged two calls. The branch lost ₹25k of annual premium and a renewal annuity that could run a decade.
The call count rated that agent fine. The branch got poorer. Blame the metric, not the agent — she was doing exactly what you measured.
This is why activity tracking quietly drains BFSI sales floors. It is gameable in an afternoon, it ignores lead quality entirely, and it sends your fastest closers a message that the busywork merchant beside them matters more.
The three metrics that actually hold up
Fair tracking starts by measuring outcomes against the opportunity that landed in each agent's queue. Three numbers do most of the work.
₹ premium recovered per agent. Not deals closed — rupees. A motor policy at ₹8k and a high-net-worth health plan at ₹2L are not the same win, and a count treats them identically. Track the premium value each agent converted against the value that was assigned to them. Now the closer who landed ₹14 lakh on fewer calls is visibly your top performer, because she is.
Response time on Grade A and B leads. Not response time on everything — that just rewards whoever clears the easy queue fastest. The leads that decide your quarter are the high-grade ones, and they have a short fuse. A Grade A buyer contacted inside the first hour converts at a multiple of one reached five hours later. So the metric is narrow on purpose: how fast did this agent reach the leads worth reaching? An agent averaging 18 minutes on Grade A leads is protecting premium. One averaging 2 hours is leaking it, even if their overall "average response time" looks tidy because they answered 40 junk leads in 90 seconds each.
Follow-up and renewal completion. Insurance is not a one-call business. The ₹40k policy that needs three touches before the buyer commits, the renewal that lapses because nobody called in the 30-day window — these are pure recoverable money sitting in process gaps. Measure whether each agent completed the follow-up sequence and closed the renewals assigned to them. A lapsed renewal is rarely the agent being lazy. It is a follow-up that slipped past its window with nothing flagging it. Fix the flag and you fix most of the lapses.
Those three together answer the only question that matters at a branch: did this agent turn the opportunity we handed them into money, and did they move fast on the parts that were time-sensitive?
How grading makes the queue fair before the day starts
You cannot measure response time on high-grade leads if nothing grades the leads. That is where most branches stall — the manager wants to track speed on the good prospects but has no objective definition of "good".
Leadkaun grades every incoming lead A through F in under 500ms — on source, product fit, premium band, declared intent, and behaviour. A ₹2L health enquiry from a buyer who opened the quote twice grades A. A six-week-old cold form grades E. The Priority Queue then orders each agent's day by that grade, so the floater buyer surfaces above the tyre-kicker automatically. The agent isn't guessing what to work first, and you are no longer measuring them on a queue they had to sort by gut.
The Missed Opportunity Engine puts a rupee figure on what slipped: a Grade A lead untouched past its window shows up as, say, ₹2L at risk this week, attributed to the agent and the lead. That is not a scolding metric. It is the manager's early-warning system — money you can still recover if someone calls today instead of finding out at month-end that it lapsed.
And the Morning Brief lands before the floor opens: who has Grade A leads waiting, what renewals fall due this week, where the ₹ at risk is concentrated. The branch starts the day pointed at the premium, not at a call target.
The audit trail that makes 1:1s fair — and keeps you compliant
Here is where fair tracking and BFSI compliance turn out to be the same project.
Sales Rep Tracking records, per agent, what actually happened: which leads they were assigned, response time on each grade, ₹ recovered, follow-up and renewal completion, and a timestamped log of calls and WhatsApp messages. That last part matters twice over. For the 1:1, it means you and the agent look at the same screen. The conversation stops being "your numbers are down, why" — which any agent can deflect — and becomes "your Grade A response time slipped to 94 minutes in week 3, and here's the ₹1.5L motor-fleet lead it cost us". Specific. Situation-first. Hard to argue with because it isn't an opinion.
For compliance, the same trail answers the questions a BFSI audit asks: who contacted whom, when, on which channel, and what was sent. Mis-selling reviews, suitability checks, and IRDAI-grade record-keeping all need that, and reconstructing it from agent memory at quarter-end is how branches get hurt. A compliance-ready audit trail protects the honest agent as much as it flags the careless one — when a complaint comes in, you can show exactly what the agent did and when.
WhatsApp belongs in that trail, not outside it. A large share of insurance follow-up — sending the quote PDF, the renewal reminder, the policy document — happens on WhatsApp. If your tracking ignores it, you are blind to most of the actual selling and your audit trail has a hole in it.
Running the 1:1 across 10 to 40 agents
A branch manager with 30 agents cannot run 30 thoughtful 1:1s if every one starts with an hour of digging for numbers. The point of all this is to walk in with the data already assembled, the same shape for everyone.
Pull each agent's period view: ₹ recovered against ₹ assigned, Grade A/B response time, follow-up completion, renewal completion, and the missed-opportunity rupees attributed to them. Five minutes of prep per agent, not an afternoon. Then the meeting is short and concrete. For the agent who closed ₹14 lakh on low call volume, the conversation is "your conversion on Grade A is the best on the floor, let's give you more of them" — not a leaderboard apology. For the agent leaking high-grade leads, it is "you're fast on the easy queue but Grade A is waiting 2 hours, that's where the ₹3L gap is this month, let's fix the morning routine".
Same data, different conversation, every agent treated by the same standard. That is what fair means in practice — not soft, just honest about where the money went and why.
The branch manager in Pune flipped his leaderboard to ₹ recovered and Grade A response time. The dialler dropped to mid-table. The closer who'd been buried at the bottom turned out to be his best agent by the only measure that pays the branch's bills. Nobody quit over it. The honest tracking just made the obvious finally visible.
If you run a BFSI sales floor, the deeper playbook on grading, queueing, and renewals lives in our BFSI use case and the insurance lead management guide. When you want to see per-agent ₹ recovered, Grade A response time, and the compliance-ready audit trail on your own pipeline, book a 15-minute demo and bring last quarter's numbers — we'll show you what fair tracking surfaces that your call counts are hiding.