Analytics toolkit skill

Analytics Toolkit

Overview

Analytics tools don't work out of the box — they work when configured with a tracking plan, proper implementation, and destinations that talk to each other. This skill covers setup and optimization across the analytics stack.

Authoritative Foundations

• Segment — CDP with 400+ destinations
• Amplitude — Product analytics, behavioral cohorts, experiment
• PostHog — Open-source analytics, session recording, feature flags
• Mixpanel — Event-based product analytics
• GA4 — Google Analytics for web + app

When to Use

Trigger phrases: "analytics setup", "Segment configuration", "Amplitude implementation", "PostHog setup", "GA4 for B2B", "analytics stack comparison"

Stage-based recommendations: references/analytics-stack-by-stage.md.
Spend tracking: gtm-spend-management vendor register.

Tool Comparison

Tool	Best For	Pricing	Key Feature
Segment	CDP — collect once, route everywhere	Free-$120/mo	400+ destinations
Amplitude	Product analytics — retention, cohorts	Free-$100K/enterprise	Behavioral cohorts
PostHog	Open-source — session replay, feature flags	Free self-hosted/Cloud $	All-in-one
Mixpanel	Event-stream analytics, simpler UI	Free-$20K/enterprise	Event explorer
GA4	Web analytics + ad attribution	Free	Google Ads integration

Implementation Pattern

// Segment → Amplitude + GA4 + Data Warehouse
analytics.track('Feature Used', { feature_name: 'Reports' });
// → Amplitude: behavioral cohort
// → GA4: custom event + conversion tracking
// → Data Warehouse: raw event for BI analysis

Implementation Checklist

• CDP implemented (Segment or Rudderstack) — single source of truth
• Server-side tracking for critical events (signup, payment)
• Client-side tracking for behavioral events
• GA4 configured with custom events + conversions
• Product analytics tool configured with retention + funnel reports
• Data warehouse receiving all events for BI

Common Pitfalls

1. Direct-to-tool without CDP. Send events directly to Amplitude + GA4 + Mixpanel. 3x the implementation. Inconsistent data. Fix: CDP as single source.
2. No server-side tracking. Ad blockers kill client-side. Critical events lost. Fix: Server-side Segment for revenue events.
3. Multiple 'source of truth' tools. Marketing trusts GA4. Product trusts Amplitude. Numbers disagree. Fix: Data warehouse as single source of truth.

Output Format

The agent delivers an analytics stack configuration plan matched to the user's stage, team, and tool question:

• Stack Recommendation: recommended combination (CDP + product analytics + web analytics) with 1-line rationale per tool choice relative to the user's company size, budget, and use case
• Implementation Priority Order: phased rollout sequence — CDP as single-collection layer first, product analytics second, web analytics third, data warehouse fourth — with reasoning for the ordering
• Event Taxonomy Sample: 5–10 prioritized events with property schema (event name, user ID, account properties, outcome it measures) aligned to the user's funnel
• Destination Routing Map: which events route to which tools (e.g., behavioral events → Amplitude for retention cohorts; conversion events → GA4 for ad attribution; all events → warehouse for BI)
• Implementation Checklist: 6-point verification covering server-side tracking for critical events, custom conversion configuration, single-source-of-truth designation, client-side redundancy, QA event audit, and stakeholder dashboard setup

Quality Check

Before delivering, verify:

• All required sections complete
• Output matches the user's stated need
• No vague or unsupported claims
• Frameworks cited where applicable

Execution Artifacts

• references/framework-notes.md — named frameworks, citation anchors, and operating assumptions
• templates/output-template.md — copy-paste deliverable structure for the user
• scripts/check-output.py — local checklist validator for required sections This skill includes lightweight artifacts the agent can load on demand:
• references/analytics-stack-by-stage.md — CDP + product + web by ARR Use the artifacts when the user asks for an implementation-ready deliverable, a repeatable workflow, or a quality check rather than generic advice.

Related Skills

• tracking-plan — Complete analytics architecture
• event-analytics — Event taxonomy and CDP design
• gtm-metrics — SaaS metrics and benchmarks