growth-context

Vivreal Growth Context Loader

When this skill activates, you have context about a growth or analytics question. Before answering with opinions or guesses, pull real data from available sources.

Trigger Keywords

This skill activates when the user mentions any of:

• Growth metrics: "conversion", "churn", "retention", "activation", "engagement"
• Traffic: "traffic", "visitors", "sessions", "page views", "bounce rate"
• Funnel: "funnel", "drop-off", "onboarding", "signup rate"
• Users: "signups", "new users", "DAU", "WAU", "MAU", "active users"
• Acquisition: "acquisition", "CAC", "customer acquisition cost", "organic", "paid traffic"
• Revenue: "MRR", "ARR", "upgrade rate", "paid conversion", "tier distribution"
• Analytics: "analytics", "metrics", "dashboard", "KPIs", "how are we doing"

Available Data Sources

1. Google Analytics 4 MCP (google-analytics)

• Traffic data, user sessions, acquisition channels, geography, device breakdown
• Use for: traffic volume questions, source attribution, landing page performance
• Tools: run_report, run_realtime_report, get_account_summaries

2. PostHog MCP (posthog)

• Product events, funnels, feature flags, session recordings, user properties
• Use for: product usage questions, funnel completion, feature adoption
• Tools: Analytics queries, feature flag status, error tracking

3. MongoDB MCP (mongodb) — already configured

• Business objects: groups, users, sites, collections
• Use for: hard business metrics (signups, groups created, tier distribution, sites deployed)
• Database: vivreal (mainDb) for cross-tenant data

4. Sentry MCP (sentry) — already configured

• Error rates, crash-free session rate, performance data
• Use for: correlating error spikes with drop-offs

Activation Procedure

1. Identify what metric the user is asking about — map their question to a specific data source
2. Pull real data first — query the relevant MCP before answering
3. Cross-reference when possible — e.g., GA4 traffic + MongoDB signups = signup conversion rate
4. Provide context — compare to previous period, note trends
5. Be honest about gaps — if an MCP server isn't configured, say so and suggest setup

Key Vivreal Metrics & Where They Live

Metric	Source	How to Calculate
Visitor → Signup rate	GA4 sessions ÷ MongoDB new users	Cross-source
Signup → Activation rate	MongoDB users with group created in 24h ÷ total new users	MongoDB
Activation → Deployment rate	MongoDB sites deployed ÷ groups created	MongoDB
Free → Paid conversion	MongoDB groups with tier != 'free' ÷ total groups	MongoDB
Tier distribution	MongoDB groups aggregated by tier	MongoDB
Feature adoption	PostHog events per feature	PostHog
Error impact on conversion	Sentry error rate vs PostHog funnel drop-off	Cross-source
Organic vs Paid quality	GA4 source/medium + MongoDB signup-to-activation by cohort	Cross-source

Common Questions & How to Answer

"How are we doing?" → Run /growth-report command. Don't guess.

"What's our conversion rate?" → Ask: "Which conversion? Visitor→signup, signup→activation, or free→paid?" Then query the right source.

"Why are signups down?" → Check GA4 traffic first (is traffic down or is conversion down?), then PostHog for signup funnel errors, then Sentry for crashes on the signup page.

"Should we invest in ads?" → First check organic conversion rate (GA4 + MongoDB). If organic converts well, ads will likely work. If organic doesn't convert, fix the funnel first.

Rules

• Never guess metrics — always query real data. If data isn't available, say "I'd need GA4/PostHog connected to answer that accurately."
• Always include time context — "signups are at X" is useless without "up Y% from last week"
• Recommend the right command — if the user's question would be better answered by /growth-report or /funnel-analysis, suggest that instead of doing ad-hoc queries
• Respect MongoDB safety rules — read-only, no groupName queries, redact credentials