NIFTY 5023,847.20-0.42%
INDIA VIX16.50+2.1%
AVG TRAIL83 BPSFY26
SIP STOPPAGE76%MAR 26
Live /
REQ // Retention Intelligence Terminal
Production

Outcomes

What the broker desk is teaching us

This view turns messy advisor behaviour into product evidence: saves, losses, false positives, no-responses, and model disagreements.

11 touched · 11 outcomes logged · 4 saved or partially saved

Saved AUM

₹1.53 Cr

₹1,27,109/yr trail protected

Lost AUM

₹15.1 L

1 hard loss logged

Follow-ups

5

3 in progress · 2 no response

AI challenged

4

Broker changed the model action

False positives

1

Human context says not a churn case

What this teaches the product

Broker override is not noise

₹80.8 L

4 AI calls were changed by the desk. Saved/partial AUM after those overrides is a direct product signal.

Silent churn is a separate workflow

2

No-response clients should not get the same long review script. They need short, low-friction prompts.

False positives still help

1

When a broker marks medical cash need or family emergency, the model learns what not to escalate next time.

Loss notes are uncomfortable but valuable

₹15.1 L

The loss reasons tell us whether the product was late, too generic, or missing a relationship cue.

Mission outcomes

Save5