How TalentGum automated 95% of support with Sagepilot
Live classes for kids in 40+ countries: public speaking, chess, coding, dance, music. 50,000+ families and 62,000+ messages a month, answered by Ananya, the AI employee, across WhatsApp, Instagram, Messenger, web chat, email, and voice.
98%
AI resolution rate
3s
average response time
2
people on support, down from 20The support team went from 20 people to 1 or 2, who now handle escalations only.
Challenge
62,000+ messages a month across WhatsApp, Instagram, and web chat, 80% of team time going to the same repeating questions, and international parents waiting hours during off-peak coverage.
Solution
Ananya, an AI employee running full pre-sales with regional pricing and demo booking, then the complete post-purchase lifecycle against real policy windows, across chat, social, email, and voice.
Results
- 98% AI resolution at a 3-second average response
- Support went from 20 people to 1 or 2 on escalations
- 24/7 coverage across every timezone
Live classes, and parents in every timezone
TalentGum teaches children aged 5 to 14 real-world skills through live online classes with real teachers: public speaking, debate, chess, coding, keyboard, dance, and art, across 40+ countries.
Every class comes with a parent who has questions, in their own timezone.
62,000 messages a month, all following patterns
At 62,000+ messages a month across WhatsApp, Instagram, and web chat, the support team spent 80% of its time on questions that followed the same pattern every time: course fit, demo scheduling, rescheduling, login issues, session policies.
Hiring more agents didn't scale, and international parents waited hours during off-peak coverage.
From lead layer to the whole operation
Sagepilot started as a WhatsApp lead-qualification layer and earned its way into the whole operation: full pre-sales conversations with regional pricing and demo booking, then complete post-purchase support.
Ananya identifies parents by their WhatsApp number, pulls the right child's profile, and manages cancellations, rescheduling, holds, and teacher transfers against real policy windows.
None of that works without real system access. Because Ananya is wired into TalentGum's class scheduling and student records, a cancellation actually executes: the session is freed, the credit carries forward, and the demo lands with the right specialist, with regional pricing applied for India, the Middle East, and the US.
She is live across WhatsApp, Instagram, Messenger, web chat, and email, and now on the phone line too, so a parent who would rather call than type reaches the same teammate, with the same child's profile in front of her.
“The AI doesn't feel like a chatbot. Parents tell us it feels like texting someone on the team.”
Twenty people to two
Support went from 20 people to 1 or 2 handling escalations only.
- Response times dropped from 15-45 minutes to under 10 seconds
- Coverage went from IST business hours to 24/7 across every timezone
- Policy enforcement became consistent, with empathetic exceptions escalated to humans
Class changes handled at 11:47 PM
Ananya identifies the parent by WhatsApp number, pulls the right child's profile, and executes cancellations, rescheduling, holds, and teacher transfers against real policy windows. This exchange is from the published study, verbatim.
- Session credit carried forward
From first message to booked demo
Course recommendations across 10 subjects, tailored by age and goals, with regional pricing for India, the Middle East, and the US, and demo booking handled end to end.
- Child age, interests, and goals captured
- Regional pricing detected: Middle East
- Demo booked with a specialist
Policy enforced, exceptions escalated
The 45-minute cancellation window applies consistently, and genuinely upset parents get a human: escalations arrive as detailed tickets with full context and a 24-hour response commitment.
- Policy window validated
- Frustrated parent detected, tone adjusted
- Escalated with full context for case review
The same teammate, now on calls
Ananya answers the phone line too. She recognises the parent, pulls the right child's profile, and executes the change on the call itself, with the confirmation following on WhatsApp so nothing depends on anyone's memory.
Meera
Sagepilot AI · Ananya01:12
Going live
Phase 1 · Lead qualification
A WhatsApp layer collects child age, course interest, and location, then routes warm leads to admissions.
Phase 2 · Full pre-sales
Course recommendations across 10 subjects, regional pricing for India, the Middle East, and the US, and end-to-end demo booking.
Phase 3 · Post-purchase support
Ananya manages the full lifecycle: cancellations with credit-back, rescheduling, holds, teacher transfers, and live-class support.
More stories
Join the brands that run their customer experience on Sagepilot
Live in about 48 hours, across every channel your customers use.