| name | analytics-funnel-attribution |
| description | Track acquisition and retention funnels for skill usage and workflow completion |
| risk | medium |
| source | .agents-phase6 |
| date_added | "2026-03-29T00:00:00.000Z" |
| usage | Use when implementing Phase 6 scale, interoperability, and analytics initiatives |
| avoid | Do not deploy without compatibility checks, rollback strategy, and measurable SLOs |
| mandates | ["define measurable success criteria and failure budgets","include platform-specific acceptance gates","include data privacy and governance checks"] |
| response | assess baseline, design minimal implementation, validate measurable targets, document rollback plan |
analytics-funnel-attribution
Overview
Track user acquisition, engagement, and retention funnels for game features—measuring skill adoption, workflow completion, and feature impact on business metrics. This skill enables data-driven iteration on framework quality and community engagement.
Key Capabilities
1. Funnel Tracking
- Event Stream: Track skill usage, workflow starts/completions, feature interactions
- Session Management: Identify user cohorts and session lifecycles
- Retention Curves: Measure D1, D7, D30 retention for features
- Convertion Tracking: Skill discovery → adoption → maintenance pipeline
- Churn Analysis: Identify users at risk and feature abandonment patterns
2. Attribution Modeling
- Multi-Touch Attribution: Credit skill exposure across multiple touchpoints
- First/Last-Click Models: Simple and complex attribution for causal inference
- Cohort Analysis: Compare feature adoption across user segments
- A/B Testing Framework: Controlled experiments on skill variants
- Incrementality Testing: Measure true impact vs. correlation
3. Analytics Infrastructure
- Event Validation: Schema validation, deduplication, late arrival handling
- Real-time Dashboards: Live funnel metrics with <5min latency
- Batch Processing: Daily/weekly cohort analysis and retention calculations
- Data Privacy: GDPR compliance with anonymization and consent tracking
- Sampling: Efficient large-scale processing via intelligent sampling
4. Insights & Reporting
- Automated Alerts: Anomaly detection on key metrics (drop >20%)
- Cohort Insights: Generation of insights via statistical testing
- Custom Reports: Self-service dashboards for different stakeholder needs
- Export APIs: CSV/JSON export for external analysis
- Predictive Models: Churn prediction, lifetime value estimation
Implementation Pattern
namespace SkillAnalytics {
public class SkillEventTracker {
public void TrackSkillDiscovery(string skillId, string userId, string source) {
var evt = new Event {
EventType = "skill_discovered",
SkillId = skillId,
UserId = userId,
Source = source,
Timestamp = DateTime.UtcNow
};
EventStore.Log(evt);
}
public void TrackSkillAdoption(string skillId, string userId, string workflowId) {
EventStore.Log(new Event {
EventType = "skill_adopted",
SkillId = skillId,
UserId = userId,
WorkflowId = workflowId,
Timestamp = DateTime.UtcNow
});
}
public async Task<FunnelMetrics> ComputeFunnel(
string skillId,
TimeSpan period,
CancellationToken ct = default)
{
var discovered = await EventStore.Count(
eventType: "skill_discovered",
skillId: skillId,
period: period
);
var adopted = await EventStore.Count(
eventType: "skill_adopted",
skillId: skillId,
period: period
);
var completed = await EventStore.Count(
eventType: "workflow_completed",
skillId: skillId,
period: period
);
return new FunnelMetrics {
Discovered = discovered,
Adopted = adopted,
AdoptionRate = (double)adopted / discovered,
Completed = completed,
CompletionRate = (double)completed / adopted
};
}
}
}
Mandates
- Consent-First: Track only GDPR-compliant events with explicit user consent
- Privacy by Design: No PII in event stream, anonymized user IDs
- Accuracy Thresholds: Metrics must be 99%+ accurate for decision-making
- Latency Targets: Real-time dashboards update within 5 minutes
- Retention Policy: Raw events deleted after 90 days, aggregates kept for 2 years
Best Practices
- Event Schema First: Design events before implementation, validate strictly
- Track Behaviors, Not Identities: Focus on what users do, not who they are
- Correlation ≠ Causation: Use proper statistical testing for causal claims
- Sample Judiciously: Balance accuracy with cost through intelligent sampling
- Iterate on Metrics: Review and refine KPIs quarterly based on learnings
Key Metrics
| Metric | Target | Alert Threshold |
|---|
| Skill Discovery → Adoption | >30% | <20% |
| Workflow Completion Rate | >80% | <60% |
| Feature Retention (D7) | >50% | <30% |
| Event Processing Latency | <1min | >5min |
| Data Freshness | <5min | >15min |
Resources