domains / marketing / analytics-tracking-setup / analytics

analytics

Skill card

combined

The three copies are not reproductions of one procedure. The microsoft/power-platform-skills copy is a Process Mining routing skill that calls native functions like get_processes and get_process_overall_metrics_v2 and routes to specialist skills. The coreyhaines31/marketingskills copy is a marketing analytics tracking skill covering GA4, GTM, UTM parameters, event naming, and measurement principles. The sickn33/agentic-awesome-skills copy is a truncated mirror of the coreyhaines31 skill, missing the later sections on GA4, GTM, UTM, and debugging. Because the first copy addresses a completely different domain (process mining) and the other two address marketing analytics tracking, crowning one would hide a real alternative. The two marketing copies overlap but the coreyhaines31 version is more complete, while the sickn33 version is a truncated mirror.

Repositories

Other copies

Combined skill

# Analytics

This skill covers two distinct domains. Determine which applies before proceeding.

## A. Process Mining Analytics (microsoft/power-platform-skills)

Use when a Process Mining overview, initial process selection, or an analytical follow-up needs routing.

1. Reuse the selected process and fresh metadata. Otherwise call `get_processes {}`, resolve the returned `P` (ask on duplicates), then `get_process_details_v2 {processId:P}` when metadata is missing. An empty process list stops analysis.
2. Call `get_process_overall_metrics_v2 {S,filterOptions:F}`. Omit absent filters. Report returned case/event/variant counts and mean `CaseDuration`/`CaseWaitingTime`; preserve native units. Missing metrics are not zero.
3. Only for requested path concentration: `get_variants_with_metrics_v2 {S,filterOptions:F,itemsPerPage:5,itemsToSkip:0,metricToSortBy:"CaseCount",sortOrder:"Descending"}`.

Route follow-ups to specialist skills: investigate-process, analyze-performance, analyze-variants, analyze-drivers, compare-cohorts, derive-metric, analyze-objects.

## B. Marketing Analytics Tracking (coreyhaines31/marketingskills)

Use when the user wants to set up, improve, or audit analytics tracking and measurement (GA4, GTM, UTM, conversion tracking, event tracking).

### Initial Assessment
Check for product marketing context first (`.agents/product-marketing.md` or `.claude/product-marketing.md`). Understand business context, current state, and technical context.

### Core Principles
1. Track for decisions, not data.
2. Start with the questions.
3. Name things consistently (object-action format, lowercase with underscores).
4. Maintain data quality.
5. Compute, don't eyeball: derive every total from row-level data; cross-check headline figures.
6. Match observation windows: rate metrics divide matching closed windows.
7. Recommend against the stated goal.

### Tracking Plan
Structure: Event Name | Category | Properties | Trigger | Notes.

### Essential Events
Marketing site: cta_clicked, form_submitted, signup_completed, demo_requested.
Product/App: onboarding_step_completed, feature_used, purchase_completed, subscription_cancelled.

### GA4 Implementation
1. Create GA4 property and data stream.
2. Install gtag.js or GTM.
3. Enable enhanced measurement.
4. Configure custom events.
5. Mark conversions in Admin.

### Google Tag Manager
Use Tags, Triggers, and Variables. Push events to the data layer.

### UTM Parameters
Standard: utm_source, utm_medium, utm_campaign, utm_content, utm_term. Lowercase, consistent separators, document all UTMs.

### Debugging and Validation
Use GA4 DebugView, GTM Preview Mode, and browser extensions. Validate events fire on correct triggers, properties populate correctly, no duplicates, works across browsers and mobile.