Engagement Analytics
Intro to Cognitive Science · Section 002 · 8-week pilot
This week4 wks8 wks
North Star
Weekly High Engagement Minutes / Class Session
41.2min▲ 45% · 8 wks
AHA-onAHA-off baseline
The delta between AHA-on and AHA-off sessions, normalized per session so a 200-student lecture and a 40-student seminar compare fairly.
Alert latency▼ 22% faster
18s
Onset of disengagement → instructor alert
Alert-to-action▼ 15% faster
41s
Alert shown → instructor response
Pulse-alert agreement▲ +9 pts
82%
% of alerts confirmed by a student's Pulse
Recovered minutes▲ vs. no action
+6.3
Engaged minutes regained / intervention
Scale gateFalse positives 11.4% · target < 15%Student consent 78% · target > 70%Renewal intent 2 / 3 pilots · target ≥ 2
Instructor retention
By week — solid = actual, dashed = target.
100%
Week 1
78%
Week 4
73%
Week 8
3.4/wk
Repeat
Feature adoption
Bar = current use · marker │ = Month-6 target.
Real-time dashboard96%
Private ping58%
AI Copilot suggestions54%
Live Student Pulse47%
Cameras-off fallback38%
Post-session summary71%
Session journey · metric at each step
Before class
Consent shown
Join + opt-in
78% consent
Live monitoring
18s alert latency
Alert → Ping
58% acted on
Copilot recovery
54% accepted
Following week
74% return
North Star chosen over attendance or logins because it reflects genuine engagement, not presence. Trust and fairness are guardrails from day one, not an afterthought at scale.