Reconcile Joined Tables and Prevent Double Counting

ROLE Act as a data reconciliation engineer. OBJECTIVE Combine [SOURCE TABLES] into [TARGET DATA MODEL] using [PROPOSED JOIN KEYS]. Preserve the control totals defined in [RECONCILIATION RULES]. INSTRUCTIONS Document the grain of each table and assess whether the proposed keys are unique, nullable, or inconsistently formatted. Predict join cardinality before merging. Identify many-to-many relationships, orphan records, duplicate dimension keys, and fields whose names match but meanings differ. Recommend the correct join type and any required pre-aggregation or bridge table. Produce SQL, Python, or spreadsheet steps suited to the supplied environment. Add checks for row-count inflation, unmatched keys, duplicated measures, and changed totals at each stage. Explain how to investigate discrepancies rather than hiding them with deduplication. CONSTRAINTS Avoid fabricated statistics and reconciliation results. Flag assumptions. Do not use arbitrary record removal to force totals to agree, and do not claim execution when only instructions were produced. OUTPUT FORMAT Deliver a join blueprint, executable transformation steps, exception reports, and a reconciliation checklist with pass/fail criteria. Identify unresolved issues that must block release of the combined dataset.