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
Learning cases
Where human judgement changed the story
Client #0115
INV_0115 · ₹6,51,315
Client replied with thumbs up but not confirmed yet.
AI Save → Desk Rebalance
₹6.5 L
Client #0486
INV_0486 · ₹55,80,804
Relationship saved and possible fresh 3L coming next week.
AI Save → Desk Grow
₹55.8 L
Client #0586
INV_0586 · ₹60,18,309
He still wants 8-10L cash out. But agreed not to shut the account. Partial save.
AI Save → Desk Rebalance
₹25.0 L
Client #0509
INV_0509 · ₹65,52,580
False positive for churn. Redemptions are medical cash need.
AI Save → Desk Check-in
Client #0934
INV_0934 · ₹15,09,783
Client already shifted most of the book to cousin's direct platform. Lesson: no quarterly review for 8 months hurt us.
AI Save → Desk Save
₹15.1 L
Follow-up debt
Open loops that still need a human nudge
Client #0374
INV_0374 · ₹19,34,048
28 Apr: Try one WhatsApp voice note today. No more than one ping.
AI Save → Desk Save
Client #0115
INV_0115 · ₹6,51,315
28 Apr: Send simple 3-line rebalance suggestion on WhatsApp.
AI Save → Desk Rebalance
₹6.5 L
Client #0597
INV_0597 · ₹5,32,340
29 Apr: Send FD vs SWP note and call after lunch. Keep it simple, she hates long PDFs.
AI Save → Desk Save
₹5.3 L
Client #0757
INV_0757 · ₹23,65,791
29 Apr: Ask for 12 min call, not full review. Mention tax only.
AI Save → Desk Save
Client #0586
INV_0586 · ₹60,18,309
30 Apr: Prepare switch list, keep 20L parked, redeem only cash need.
AI Save → Desk Rebalance
₹25.0 L
Client #0893
INV_0893 · ₹60,72,682
30 Apr: Call after salary credit. She said month-end, so don't call before 30th.
AI Save → Desk Save
₹42.5 L
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
Broker notes
The messy bits worth preserving
Client #0729
INV_0729 · ₹67,28,911
Call went better once spouse joined. They don't want jargon. Keep it family-goal focused. She said okay to pause redemption.
AI Save → Desk Save
₹60.6 L
Client #0115
INV_0115 · ₹6,51,315
Client said send on WA, he will read at night. Not a phone-call person. Don't force call.
AI Save → Desk Rebalance
₹6.5 L
Client #0324
INV_0324 · ₹21,40,883
Client already did one redemption. Daughter is sensible, asked for Saturday review. I should have included family earlier.
AI Save → Desk Save
₹11.8 L
Client #0486
INV_0486 · ₹55,80,804
Client was laughing about markets, not scared. She wants a bonus deployment idea. I nearly over-explained; keep next msg short.
AI Save → Desk Grow
₹55.8 L
Client #0597
INV_0597 · ₹5,32,340
Called client. She said 'not angry, just tired of guessing'. wants FD comparison. I think relationship can be saved but need to be direct about last redemptions.
AI Save → Desk Save
₹5.3 L
Client #0586
INV_0586 · ₹60,18,309
Client was irritated, not gone. Said 'show me what you will change'. Need a cleaner switch plan. Don't defend old fund.
AI Save → Desk Rebalance
₹25.0 L