Admin view

Churn detection agent

An AI agent that monitors all new advisors before their first booking, scores their risk level based on schedule adherence, and delivers personalised resource recommendations automatically.

How the agent works
01
Data ingestion
  • Days since last portal login
  • Schedule adherence % (sessions completed on time)
  • Quotes sent
  • Bookings made
  • Days since joining Fora
  • Track assignment (A / B / C / D)
  • Calendar sessions marked complete
02
Risk scoring
  • Goal: first booking within 60 days
  • ─── Days 1–20: departure window ───
  • No risk for zero bookings — too early
  • 7+ days no login + <10% training → medium
  • 10+ days no login → medium
  • ─── Days 21–45: activation window ───
  • Login gap 10d + no quotes + <30% training → high
  • Login gap 7d OR no quotes after day 30 → medium
  • ─── Days 46–55: final approach ───
  • No quote ever sent → high
  • Quotes sent, no booking → medium
  • ─── Days 56–60: critical window ───
  • Gone dark OR no quotes → high
  • Active with quotes → medium
  • Booking made → always low
03
What it triggers
  • Low risk → no action, continue monitoring
  • Medium risk → track-matched resource pack surfaced
  • High risk → resource pack + escalation flag
  • Resources matched to advisor's barrier (A/B/C)
  • Three support paths offered: Sidekick · Forum · HQ
  • Outreach email drafted automatically
  • All actions logged for Fora team review
Risk tiers
Low risk

Advisor is active and on trajectory — or has already made their first booking. In the first 20 days, almost everyone is low risk. No intervention needed.

Medium risk

Engagement is slowing or a key milestone is overdue for their window — e.g. no quotes by day 30, or login gaps appearing. Agent surfaces track-matched resources.

High risk

Requires BOTH a time signal AND disengagement signal in their window — e.g. 10+ day login gap with no quotes and low training. In days 46–60, no quote sent at all triggers high risk.

Live simulation
6 mock advisors with real behavioral signals — click Run agent to watch it work
7
Advisors monitored
2
High risk
1
Medium risk
0
Resources surfaced
MR
Maya Rodriguez
Track A — client acquisition · Day 22
9d no login0 quotesSchedule 30%0 bookings
high risk
JO
James Okafor
Track B — platform confidence · Day 18
3d no login1 quotesSchedule 80%0 bookings
low risk
On track — no intervention needed.
SC
Sophie Chen
Track C — pricing confidence · Day 35
4d no login3 quotesSchedule 90%0 bookings
low risk
On track — no intervention needed.
DP
David Park
Track A — client acquisition · Day 41
10d no login0 quotesSchedule 15%0 bookings
high risk
PN
Priya Nair
Track B — platform confidence · Day 12
1d no login2 quotesSchedule 75%1 booking
low risk
On track — no intervention needed.
TA
Tom Alvarez
Track C — pricing confidence · Day 28
5d no login5 quotesSchedule 100%0 bookings
low risk
On track — no intervention needed.
EV
Elena Vasquez
Track D — supplier relationships · Day 32
8d no login2 quotesSchedule 55%0 bookings
medium risk