New: Cookbooks and AI ExplanationsStep-by-Step recipes to solve problems connected to Roadmaps and Cheat Sheets. Need more details? Use AI buttons for structured and simple explanations with concrete examples throughout the whole platform.Take a look
A stakeholder says the numbers look off, and all you have is a guess about where to start looking.
What you'll have at the end
A short root-cause note that names the exact upstream table or job that broke, distinct from the downstream number where the symptom showed up
You need
A data warehouse table feeding a dashboard panel, built through at least one modeled table (for example a transformation-tool model), with query access to that table and everything feeding it, plus a stakeholder's specific complaint about one number.
Not covered
Confirming that the number is a definitions mismatch, like the stakeholder's own tool grouping by a different date field, since that's a conversation about what the metric means, not a broken pipeline to trace.
Flag a field that changed meaning instead of shape
when a number has been quietly wrong for weeks with nothing broken and no fresh change to trace, not freshly wrong right after one edit
Trace one field from source to dashboard
when you're about to change a column and need to know what it would break, before anything is actually broken yet
Checked 26 Aug 2026
Part of the Data Observability cookbook