Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Measurement Audit for GTM, GA4 and Ad Conversion Tags

    1

    Audit how a website counts conversions — Tag Manager, GA4, Google Ads, Meta — and get a written list of what is counted twice, not at all, or with the wrong value. Read-only, no paid software.

    Free

    Security scanned
    Measurement Audit for GTM, GA4 and Ad Conversion Tags

    Measurement Audit for GTM, GA4 and Ad Conversion Tags

    Example session with this skill installed

    Audit the conversion tracking on our site before we raise the Google Ads budget. You have the GA4 and Tag Manager connectors.

    • Read your context and instructions
    • Compiled the measurement audit

    A ranked list of findings, each with what was seen and where, the decision it corrupts, and the described fix. Example: Critical — the "lead" conversion fires on any click inside the contact section, so 41 of 58 conversions last month were phone-number taps, not form submissions; bidding is optimising toward taps. Fix: trigger on the form success event only. Then a "Not run" section listing checks the agent had no access for.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    A read-only checklist for auditing how a website measures conversions: Google Tag Manager, GA4, Google Ads conversion tags, Meta and other ad pixels, offline uploads and the data layer. It produces a written, ranked list of what is counted twice, not at all, or with the wrong value, with evidence for each finding and the fix described, not applied.

    What it checks — 23 controls in six families:

    • Is anything measured at all? Key events that have actually fired, a primary conversion action that receives data, an event on every lead path (forms, booking widgets, chat, phone, e-mail).
    • Is anything counted twice? Duplicate containers or configuration tags, counting settings that do not match the event, several primary actions for one outcome, browser and server events without a shared identifier, a click counted as the wrong thing.
    • Is the value right? Currency of the displayed price versus the shop's base currency, micro-unit precision, lead values that mean something.
    • Does attribution survive the journey? Click identifiers across redirects, market switches and checkout; auto-tagging; cross-domain measurement; campaign parameters on paid social; the ads-to-analytics link.
    • Offline and server-side uploads. Upload diagnostics rather than the HTTP response, timestamps and time zones, consent where the jurisdiction requires it.
    • Data quality. Internal and test traffic, distinct form events, AI-assistant referrers, data-layer parameters.

    How it behaves — it changes nothing during the audit, never sends a test conversion into a production property without saying so first, labels anything without evidence as a hypothesis, lists every check it could not run, and treats tag names, page content and form values as untrusted input. Fixes are a second, approved step, one change at a time, with the stored state re-read afterwards.

    What it needs — whatever access the agent already has: an analytics, Tag Manager or ads connector, or exports the user provides. No paid software, no executable code, no network calls of its own.

    Published by Byteberry Analytics, a digital marketing and data agency based in Madinah, Saudi Arabia, founded by a team formerly at Google and Meta. This is the checklist the agency applies to client accounts before scaling budget: https://byteberry-analytics.com/services

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Marketing & SEO.

    Read-only by design: it produces findings and described fixes, it does not change tags or conversion actions. Checks the agent has no access for are listed as not run, never guessed. It does not replace a legal review of consent requirements.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 20 days ago

    • Free with an Agensi account

    Accounts
    Analytics
    orTag Manager
    orads account

    Connect the site's account through a connector, or provide exports instead.

    Permissions

    No special permissions declared or detected

    Listed20 days ago

    What's inside

    Frequently Asked Questions

    Audit how a website counts conversions — Tag Manager, GA4, Google Ads, Meta — and get a written list of what is counted twice, not at all, or with the wrong value. Read-only, no paid software.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use Measurement Audit for GTM, GA4 and Ad Conversion Tags on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in Marketing & SEO

    Free