Skip to content
Validation

Data Parity

Prove two datasets match after a migration, refactor, or rewrite.

Validate that two tables or query results are identical or diagnose exactly how they differ.

Validate data changes with evidence, not hope.

The problem it solves

Most migration sign-offs are vibes. You need row counts, schema comparison, key coverage, and targeted diffs before you can actually trust the new result.

What it does

Compares two tables or query results.
Supports safer profile-based comparison for sensitive or large data.
Identifies row count, schema, key, and value differences.
Produces evidence that helps teams validate migrations.

When to use it

AFTER

a warehouse migration

AFTER

an ETL or dbt refactor

WHEN

replacing one table or query with another

BEFORE

signing off on data parity in critical workflows

How it works

The skill inspects both sides, confirms keys and exclusions, profiles cheaply first, and runs targeted row-level diff only when appropriate.

Tools used

data_diff

Compare two tables or queries with profile or row-level diff strategies.

schema_inspect

Inspect a table's columns, types, nullability, and constraints.

sql_execute

Run SQL against a configured warehouse and return a formatted result set.

Related skills