AI Blocks › Track your numbers
GA4 raw data analyst
Point it at your GA4 export in BigQuery. Get sessions, channels and sign-ups, plus why they differ from the GA4 screen.
GA4 raw data analyst is a free AI skill for the job "Track your numbers". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · MIT licence · Made by wizard7558 · Level: expert · Thinking: high
Made by wizard7558. See the original on GitHub; we copied it as it is, with its licence.
What it does
Your GA4 numbers finally match what you can trace. Sessions, channels and sign-ups come straight from the raw daily export in BigQuery, with a clear note on why they differ from the GA4 screen. It starts by fixing the boundary: which dataset, which dates, which events count as a key event. Then it checks the table layout before it writes a single query. Ready-made SQL covers sessions, channels, landing pages, traffic source checks, ecommerce and a screen reconciliation. Every query is shown first, with a cost estimate and a byte cap. Late events are flagged, since daily tables can still change for a few days.
You give it
- Access to your GA4 BigQuery export
- Dataset name and billing project
- Date range to cover
- List of key events (optional)
You get
- Session and channel tables from raw events
- Landing page and source comparison
- Key event and ecommerce counts
- A note explaining gaps against the GA4 screen
Use it when
- Your GA4 report and your own sign-up count do not agree and you need to know why.
- You want sessions by channel from raw export tables instead of the GA4 screen.
- You need landing page and UTM checks for a set of dates.
- You want purchase or lead events counted the same way every month.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run SQL on your own Google BigQuery data.