Skip to main content
Before diving into analytics, it helps to know whether your data is complete and reliable. The AI Analyst includes built-in diagnostic checks that surface issues proactively.

Two Types of Health Checks

CheckWhat It Answers
Data HealthIs my data fresh and available? Are tables up to date?
Attribution HealthIs my tracking working? Are marketing sources being captured?
These checks address different concerns:
  • Data Health is about your data pipeline — whether data is flowing from source systems to your warehouse
  • Attribution Health is about your tracking setup — whether UTM parameters and touchpoints are being captured correctly
A table can be perfectly fresh but still show poor attribution. Conversely, tracking can be excellent but data stale. These are independent issues with different solutions.

When to Run Diagnostics

Before starting an analysis: Ask “How is my data health?” to confirm the tables you need are current. When results look wrong: If numbers seem off, check Data Health first (is the data stale?), then Attribution Health (is tracking capturing sources correctly?). Proactively: Run diagnostics periodically to catch issues before they affect decisions.

Diagnostic Reports


Quick Reference

QuestionDiagnosticWhat You Learn
”How is my data health?”Data HealthTable freshness, which domains are ready
”Which tables are fresh?”Data HealthSpecific table-level status
”How is my attribution health?”Attribution Health% of orders with attribution, tracking gaps
”Why is traffic showing as direct/none?”Attribution HealthCauses of missing attribution