Diagnosing Which Data Quality Dimension a Symptom Belongs To

A finance analyst reports: "the total_revenue number on the weekly report has been off for a while — I'm not sure since when, or by how much, but it doesn't match what I'd expect." You have no other information yet. The total_revenue figure is computed by a dbt model that sums orders.amount joined against a customers dimension table.

1. List at least four distinct root causes this symptom could have, each corresponding to a different dimension of data quality (accuracy, completeness, timeliness, consistency, uniqueness), and give a concrete mechanism for each.
2. For each of the four, describe the specific query or check you would run first to confirm or rule it out, in an order that maximizes information gained per check.
3. Explain why starting with "let's just re-run the pipeline and see if the number changes" is a weak first move, even though it's often the fastest thing to try.