analytics-tracking

Overview of analytics-tracking skill

What analytics-tracking is for

The analytics-tracking skill helps an AI agent plan, audit, and debug analytics implementation work, especially for GA4, Google Tag Manager, event taxonomy, conversion tracking, UTM capture, custom dimensions, and data quality. It is best for teams that need reliable instrumentation before they trust reports, ads optimization, attribution, or funnel analysis.

Best-fit users and jobs

Use this analytics-tracking skill if you are a founder, marketer, product manager, analytics engineer, or developer trying to answer: “Are we capturing the right customer actions, with the right names and parameters, in the right tools?” It is strongest for Analytics Implementation tasks such as building a tracking plan, standardizing event names, reviewing GTM setup, diagnosing missing GA4 events, or turning business goals into measurable conversion events.

What makes it useful beyond a generic prompt

The skill includes opinionated implementation references, not just broad analytics advice. The repository provides a debugging playbook, an event taxonomy guide, GTM SaaS patterns, and a Python tracking plan generator. That gives the agent a concrete workflow: define business context, map events, apply naming rules, recommend GA4/GTM configuration, and verify data through the stack from app dataLayer pushes to GA4 DebugView.

When this skill is not the right fit

Do not install analytics-tracking if your main goal is campaign performance analysis, dashboard design, or interpreting product usage trends after data is already clean. It is about instrumentation quality. For campaign reporting, use a campaign analytics workflow; for BI or product analytics exploration, use a product analytics workflow.

How to Use analytics-tracking skill

analytics-tracking install and repository path

Install the skill from the GitHub repository with:

npx skills add alirezarezvani/claude-skills --skill analytics-tracking

The source path is marketing-skill/skills/analytics-tracking. After install, read SKILL.md first, then inspect these support files before asking the agent for implementation output:

• references/event-taxonomy-guide.md for naming and parameter standards
• references/gtm-patterns.md for GTM tag, trigger, and variable patterns
• references/debugging-playbook.md for missing-event diagnosis
• scripts/tracking_plan_generator.py for structured tracking plan generation logic

Inputs the skill needs to produce useful output

For strong analytics-tracking usage, provide implementation context instead of asking for “a GA4 setup.” Include:

• Business type, such as SaaS, ecommerce, marketplace, or lead generation
• Key pages, routes, product flows, and forms
• Primary conversions and secondary micro-conversions
• Current stack: GA4, GTM, server-side tagging, Segment, RudderStack, custom code
• Whether consent mode, GDPR/CCPA, or cookie banners affect tracking
• Existing event names, known gaps, duplicate events, or broken conversions
• Paid channels that need UTM or conversion consistency

A weak prompt is: “Set up analytics for my app.”
A stronger prompt is: “Use analytics-tracking to create a GA4 and GTM tracking plan for a B2B SaaS app with homepage, pricing, signup, onboarding, dashboard, demo request, trial start, and subscription purchase. We use GTM, need consent-aware tracking, and want event names that follow a consistent taxonomy.”

Practical workflow for implementation

Start with the event taxonomy before configuring tags. Ask the agent to define events using the repository’s object-action convention, then specify triggers, required parameters, optional parameters, conversion status, and priority. Next, map each event to a GTM pattern: preferably app-level dataLayer.push() events for important actions, rather than fragile click-only triggers.

For debugging, ask the agent to use the bottom-up stack from the playbook: app code or dataLayer, GTM firing, network requests, GA4 processing, and finally GA4 reports or DebugView. This prevents the common mistake of checking GA4 reports first and guessing why data is missing.

Example prompt that invokes the skill well

“Use the analytics-tracking skill as an Analytics Implementation guide. Audit our current GA4/GTM setup for a SaaS funnel: /, /pricing, /signup, /app/onboarding, /billing. Current events are SignUp, signup_complete, trialStart, and purchase. Problems: signup conversion is missing in GA4, pricing views are duplicated, and paid campaigns need reliable UTM capture. Produce: 1) corrected event taxonomy, 2) GTM tag/trigger/variable changes, 3) GA4 custom dimensions, 4) debugging checklist by layer, and 5) a rollout QA plan.”

analytics-tracking skill FAQ

Is analytics-tracking beginner-friendly?

Yes, if you can describe your website or product flow. The skill can translate rough business goals into events and parameters, but you still need access to GA4, GTM, site code, or the developer who controls them. Beginners should ask for a step-by-step plan and definitions for terms like dataLayer, custom dimensions, and DebugView.

How is this better than a normal GA4 prompt?

A normal prompt may produce a generic event list. The analytics-tracking skill is more implementation-oriented: it pushes the agent to standardize names, avoid duplicate events, define GTM architecture, consider consent and UTM handling, and debug from the source event upward. That matters because analytics failures are often silent and configuration-specific.

Can it generate a full tracking plan?

Yes. The included scripts/tracking_plan_generator.py indicates a workflow for producing event taxonomy, GTM configuration, and GA4 dimension recommendations from structured inputs such as business type, key pages, conversion actions, paid channels, and consent requirements. Treat the output as a draft to review with engineering and marketing stakeholders.

What are the main adoption blockers?

The biggest blockers are incomplete context, lack of tool access, and unclear business definitions. If your team has not agreed what counts as a conversion, lead, signup, trial, or purchase, the skill can propose a taxonomy but cannot resolve business ownership. Also, GTM click triggers may be unreliable unless your site has stable selectors or app-level event pushes.

How to Improve analytics-tracking skill

Improve analytics-tracking results with stronger context

The fastest way to improve analytics-tracking output is to provide real flows and constraints. Add URLs or route names, screenshots of GTM tags, current GA4 event lists, examples of broken events, consent banner behavior, and the exact conversion definitions used by sales or marketing. The agent can then produce implementation-ready recommendations instead of a theoretical tracking plan.

Common failure modes to watch for

Watch for event sprawl, inconsistent naming, missing parameters, overreliance on button-click triggers, duplicate page views in single-page apps, and conversions marked too early in the journey. Also check whether personally identifiable information could leak into GA4 parameters. Ask the agent to flag privacy risks and define which parameters should never be sent.

How to iterate after the first output

After the first plan, run a review pass with three lenses: business value, technical reliability, and reporting usefulness. Ask: “Which events are essential for decision-making?”, “Which require developer instrumentation instead of GTM-only tracking?”, and “Which parameters should become GA4 custom dimensions?” Then request a QA checklist for GTM Preview, browser network requests, and GA4 DebugView.

Repository improvements worth considering

To improve the analytics-tracking skill itself, add example input/output files for common cases such as SaaS signup, ecommerce checkout, and lead generation. A sample tracking-plan.json, a consent-mode checklist, and a concise GA4 custom dimensions template would make the skill easier to adopt. The existing references are useful; adding end-to-end examples would reduce setup ambiguity for new users.