AI Blocks › Track your numbers
Ad data quality checker
Give it your ad and sales tables. Get nine checks that catch missing spend, duplicates and empty columns.
Ad data quality checker 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
Nine checks run on your ad and sales tables and catch missing spend, duplicates, double-counted leads and empty columns before they distort a report. The skill makes you state your evidence first: the reporting window, time zone and which systems you are comparing. It then runs the checks on typed inputs, including every excluded row, so nothing disappears quietly. It checks that spend adds up across all buckets, that funnel counts add up with unknowns kept, that rates use the right population, and that each lead has one primary record. It compares sources that claim to measure the same thing. The result is complete and repeatable, with the scope of each check labelled.
You give it
- Ad and sales tables, with the date window
- The systems you want to compare
- The time zone for the report
- A list of expected columns (optional)
You get
- Results for nine quality checks
- Rows or groups that failed
- A column fill report
- The scope of each check
- A repeatable record of the findings
Use it when
- Your ad spend total does not match the platform's invoice.
- You suspect leads are counted twice in your report.
- A column in your data looks empty and you want proof.
- You need a repeatable check before sending a report to a client.
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 Node code and SQL on your own BigQuery data.