メニュー
Build a measurement strategy aligned to business goals, then implement GA4, Google Tag Manager, and event tracking with naming conventions, UTM strategy, and privacy compliance.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/heyharmony/skills --skill set-up-analytics-trackingコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
Extract every action item from your last meeting — who owns it, what's the deadline, and what's the context. Trigger on: action items, what came out of the meeting, meeting actions, follow-ups from the call, what did we agree to do, meeting takeaways, tasks from the meeting, who's doing what, meeting outcomes.
Create and optimize ad copy across platforms using conversation insights, performance data, and platform specifications
Audit and optimize content for AI search visibility across ChatGPT, Perplexity, Google AI Overviews, Claude, and Gemini. Analyze extractability, authority signals, and citation opportunities to maximize zero-click AI search presence.
| name | Set Up Analytics Tracking |
| description | Build a measurement strategy aligned to business goals, then implement GA4, Google Tag Manager, and event tracking with naming conventions, UTM strategy, and privacy compliance. |
| role | Marketing |
| action | Execute |
| connectors | ["Google Drive","Slack","Notion","HubSpot"] |
| triggerKeywords | ["analytics tracking","GA4 setup","Google Tag Manager","UTM tracking","event tracking","conversion tracking","measurement plan","tracking plan","what should I track","analytics implementation","data quality"] |
| author | Harmony |
| version | 1.0.0 |
| icon | sparkles |
| cover | |
| screenshots | [] |
| category | Marketing |
This skill helps you establish a measurement system that answers the questions driving your business: Which channels deliver value? What's our true conversion rate? Where are users dropping off? By grounding tracking in business goals—surfaced from your team's strategy discussions, product requirements, and funnel definitions—you build analytics that actually inform decisions.
We guide you through defining what matters, designing a clean tracking architecture (GA4 + GTM), naming conventions that stick, UTM strategy, and data quality safeguards. Privacy compliance (GDPR, CCPA) is built in from the start.
Align on business goals. Review conversations from leadership, product, and planning meetings where KPIs and success metrics were debated. Extract 3–5 core objectives (e.g., "drive high-value signups," "improve onboarding completion," "increase feature adoption"). Document the "why" for each; this justifies every event you track.
Define conversion events. List key user actions tied to your goals: signup, demo request, payment, feature adoption, support ticket. For each, specify trigger condition (exactly when it fires), business value (why we care), and owner (who defines/maintains it). Cross-reference HubSpot funnel definitions or Notion product requirements.
Design your GA4 property and data streams. Create separate properties for dev/staging/production to avoid data pollution. Set up data streams for web, app (if applicable). Configure consent mode if GDPR applies. Test with a dummy event before going live.
Build your GTM container. Create tags for GA4 event tracking, UTM capture, and third-party tools (LinkedIn, Intercom). Use triggers to fire tags only when conditions match (form submit, page view type, user segment). Implement data layer best practices to pass event properties reliably.
Establish naming conventions and properties. Create a shared document (Google Drive or Notion) naming all events, categories, and custom properties. Use snake_case, be specific ("email_signup_form_submit" not "signup"), include a data dictionary with allowed values. Link to this doc in Slack #analytics so everyone references the same spec.
Plan your UTM strategy. Decide on source/medium/campaign/content naming for paid, organic, and internal campaigns. Document templates for each channel (e.g., LinkedIn = "utm_source=linkedin&utm_medium=social&utm_campaign=demo-ebook-q2"). Automate UTM generation in your email platform, ad tools, or spreadsheet to avoid typos.
Deploy, validate, and iterate. Use Google Analytics Debugger or GTM's Preview Mode to confirm events fire with correct properties. Set up automated alerts for data anomalies (daily users drop 50%, conversion rate spikes). Monitor Privacy violations or incomplete data, then refine your tracking. Schedule quarterly reviews to retire obsolete events and add new ones.
Provide a Tracking Implementation Plan with these sections:
Company: Claritylabs (B2B analytics SaaS, building marketing site + product)
Business Goals (from planning call):
Conversion Events Defined:
| Event Name | Trigger | Business Value | Owner |
|---|---|---|---|
| website_form_submit | Contact form completed | Lead capture | Marketing |
| demo_request_click | "Book a Demo" button clicked | Qualified intent | Sales Dev |
| signup_complete | New account created (web) | Free user acquisition | Product |
| payment_initiated | Checkout flow started | Revenue funnel visibility | Finance |
| payment_complete | Invoice paid | Revenue | Finance |
| feature_first_use | User opens specific feature (e.g., "custom dashboards") | Feature adoption | Product |
| support_ticket_opened | Ticket submitted | Churn risk signal | Support |
GA4 Setup Checklist:
GTM Container Setup:
Container 1: "Claritylabs Marketing Website"
Container 2: "Claritylabs Product App"
Event Naming & Data Dictionary:
website_form_submit
- Properties: form_type (string: "contact", "newsletter", "demo"), form_name (string), form_position (string: "hero", "mid-page", "footer")
- Example: {event: "website_form_submit", form_type: "demo", form_name: "homepage_hero_cta", form_position: "hero"}
feature_first_use
- Properties: feature_id (string: "custom_dashboards", "alerts", "data_export"), user_segment (string: "free", "pro", "enterprise")
- Example: {event: "feature_first_use", feature_id: "custom_dashboards", user_segment: "pro"}
payment_complete
- Properties: revenue (number, in cents), currency (string: "USD"), product_id (string), transaction_id (string)
- Example: {event: "payment_complete", revenue: 99900, currency: "USD", product_id: "pro_annual", transaction_id: "txn_12345"}
UTM Strategy:
utm_source=google&utm_medium=cpc&utm_campaign=seo_alternative_q1&utm_content=claritylabs_analyticsutm_source=linkedin&utm_medium=social&utm_campaign=demo_webinar_march&utm_content=video_carouselutm_source=email&utm_medium=newsletter&utm_campaign=weekly_digest_mar22&utm_content=feature_spotlightutm_source=direct&utm_medium=internal&utm_campaign=onboardingData Quality Rules:
Privacy Compliance:
Success Metrics:
Next Steps & Timeline:
Cross-domain tracking (subdomain or separate domain): Use GA4's cross-domain measurement feature. In GTM, configure a custom parameter to identify "cross_domain_user" so conversions on different domains roll up to a single user journey.
Mobile app + web tracking: Use the same GA4 property for both (separate data streams). Align event names across platforms (e.g., both app and web fire "signup_complete" with same properties) so you can compare cohorts.
Server-side tracking (backend events): Use Measurement Protocol to send events from your server (e.g., "payment_complete" fired by billing system, not frontend). Validate that server events have matching event structure to client-side events.
High-traffic sites (sampling limit): GA4 enforces a 1M events/day limit for free tier. Monitor daily volume; if approaching, sample lower-priority events (page_view) or filter out internal traffic. Upgrade to GA4 360 if needed.
Changing requirements mid-stream: Don't delete old events; mark them deprecated in your data dictionary. Create a new version (e.g., "signup_complete_v2") and run both in parallel for 30 days to compare, then switch.
Third-party integrations (Slack alerts, Intercom, CRM): Use GTM Server-Side Tags or API connectors (e.g., Zapier) to push GA4 events to downstream tools. Test the pipeline end-to-end before going live.