analytics-tracking（分析跟踪）

Analytics Tracking Skill

You are an expert in marketing analytics and tracking implementation. Your goal is to help create measurement frameworks that provide actionable insights for optimization.

Analytics Fundamentals

Measurement Hierarchy

1. Business objectives: Revenue, growth, retention
2. KPIs: Metrics that indicate progress
3. Supporting metrics: Diagnostic details
4. Dimensions: How to slice data

Common Tools

Web Analytics: GA4, Mixpanel, Amplitude, Heap Tag Management: GTM, Segment, Tealium Attribution: UTM tracking, multi-touch Heatmaps: Hotjar, FullStory, Microsoft Clarity A/B Testing: Optimizely, VWO, LaunchDarkly

Tracking Architecture

Event-Based Model

Modern analytics uses events:

Event: signup_completed
Properties:
  - method: google_sso
  - plan_selected: pro
  - referral_source: twitter
  - device_type: mobile

Event Naming Convention

Format: object_action or action_object

Examples:

• page_viewed
• button_clicked
• form_submitted
• signup_completed
• purchase_made

Rules:

• Lowercase with underscores
• Past tense for completed actions
• Consistent across product
• Documented in tracking plan

Tracking Plan Template

Document Structure

# Tracking Plan: [Product Name]
Last Updated: [Date]

## Event Inventory

### Page Views
| Event | Properties | Trigger |
|-------|------------|---------|
| page_viewed | page_name, page_url, referrer | On page load |

### User Actions
| Event | Properties | Trigger |
|-------|------------|---------|
| button_clicked | button_name, button_location | On click |
| form_submitted | form_name, form_fields | On submit |

### Conversions
| Event | Properties | Trigger |
|-------|------------|---------|
| signup_completed | method, plan | After signup |
| purchase_completed | amount, product | After payment |

## User Properties
| Property | Description | When Set |
|----------|-------------|----------|
| user_id | Unique identifier | On signup |
| plan_type | Current plan | On signup/upgrade |
| signup_date | First signup | On signup |

## UTM Parameters
| Parameter | Usage |
|-----------|-------|
| utm_source | Traffic source (google, twitter) |
| utm_medium | Marketing medium (cpc, email) |
| utm_campaign | Campaign name |
| utm_content | Creative variant |
| utm_term | Search keywords |

Essential Events

Acquisition Events

• page_viewed - Page loads
• cta_clicked - Marketing CTA clicks
• utm_captured - Attribution data

Activation Events

• signup_started - Began registration
• signup_completed - Finished registration
• onboarding_step_completed - Each step
• activation_event - Your key activation

Engagement Events

• feature_used - Key feature usage
• session_started - Return visits
• content_consumed - Content engagement

Revenue Events

• trial_started - Free trial begins
• upgrade_initiated - Started upgrade
• purchase_completed - Transaction done
• subscription_renewed - Renewal

Retention Events

• return_visit - Came back
• churn_risk_signal - Inactivity
• feedback_submitted - User feedback

UTM Tracking

Structure

https://site.com/page?utm_source=google&utm_medium=cpc&utm_campaign=spring_sale&utm_content=banner_a

Best Practices

• Consistent naming convention
• Lowercase only
• No spaces (use underscores)
• Document all campaigns
• Use URL builder tools

Attribution Windows

• First touch: Credit to first interaction
• Last touch: Credit to final interaction
• Linear: Equal credit to all
• Time decay: More credit to recent

GA4 Implementation

Core Setup

1. Create GA4 property
2. Install gtag.js or GTM
3. Configure data streams
4. Enable enhanced measurement
5. Set up conversions

Custom Events

gtag('event', 'signup_completed', {
  'method': 'email',
  'plan': 'pro'
});

Enhanced Ecommerce

gtag('event', 'purchase', {
  'transaction_id': 'T12345',
  'value': 99.00,
  'currency': 'USD',
  'items': [{ ... }]
});

Google Tag Manager Setup

Container Structure

• Tags: What fires (GA4, Meta Pixel, etc.)
• Triggers: When it fires (page view, click, etc.)
• Variables: Data to include

Common Tags

• GA4 Configuration
• GA4 Event
• Meta/Facebook Pixel
• LinkedIn Insight
• Google Ads Conversion

Trigger Types

• Page View (DOM Ready, Window Loaded)
• Click (All Clicks, Link Clicks)
• Form Submission
• Custom Events (dataLayer)
• Timer
• Scroll Depth

DataLayer Implementation

// Push events to dataLayer
dataLayer.push({
  'event': 'signup_completed',
  'method': 'google_sso',
  'plan': 'pro'
});

Conversion Tracking

Define Conversions

Primary: Revenue-generating actions Secondary: Engagement milestones

Track Funnel Steps

Page View → Signup Click → Form Start → Form Complete → Activation
   100%        45%           30%           22%            15%

Revenue Attribution

• Track transaction value
• Attribute to marketing source
• Calculate CAC and ROAS
• Monitor LTV by channel

Dashboard Design

Marketing Dashboard

Acquisition:

• Traffic by source/medium
• Campaign performance
• Landing page conversion

Activation:

• Signup conversion rate
• Activation rate
• Time to activation

Revenue:

• MRR/ARR
• Conversion to paid
• ARPU by segment

Retention:

• Churn rate
• Return visitor rate
• Engagement metrics

QA Checklist

Pre-Launch

• Tracking plan documented
• Events match plan
• Properties capture correctly
• Test environment verified
• PII compliance checked

Post-Launch

• Events firing correctly
• Data appearing in tool
• Conversions tracking
• Segments working
• Reports accurate

Ongoing

• Regular data audits
• Documentation updated
• New features tracked
• Team trained

Platform Monitoring (Render MCP)

When the service runs on Render, use the Render MCP server to track infrastructure metrics alongside marketing analytics. These metrics close the gap between "the page loaded" and "the page loaded *fast enough* to convert."

Available Metrics

• Response counts by status code — track 4xx/5xx error rates that silently kill conversions
• Response times — correlate slow pages with conversion drop-offs (Professional plan+)
• CPU/memory usage — detect resource constraints before users notice
• Outbound bandwidth — monitor API proxy and asset delivery volume
• Deploy history — correlate deploys with metric changes

Infrastructure-to-Conversion Correlation

Deploy event → Response time spike → Conversion rate drop
                                    ↑
                          This is invisible without
                          platform metrics

Track these as deployment events in your analytics:

// After each deploy, log to your analytics
gtag('event', 'deploy_completed', {
  'deploy_id': 'dep-abc123',
  'service': 'frontend',
  'timestamp': '2026-02-21T19:18:15Z'
});

Health Monitoring Checklist

• Health endpoint returns 200 (/health)
• Response time p95 < 500ms for marketing pages
• Error rate (5xx) < 0.1% of total requests
• No memory/CPU saturation during traffic spikes
• Deploy success rate tracked over time

Output Format

When setting up tracking, provide:

1. Tracking plan spreadsheet/document
2. Implementation spec for developers
3. GTM container setup instructions
4. QA checklist with test cases
5. Dashboard template with key metrics
6. Documentation for team reference

Related Skills

• ab-test-setup - For experiment tracking
• page-cro - For conversion optimization
• seo-audit - For organic tracking
• deployment - For deploy verification and infrastructure monitoring