The project began by building a centralized BigQuery warehouse to ingest and normalize all historical Excel-based crawled data. Python scripts were created to clean inconsistent formats, extract relevant fields, and manually map UUIDs to each product.To support fresh crawl data, the vendor transitioned to pushing files directly into sharded BigQuery tables. This enabled the creation of scheduled queries and stored procedures that performed automatic table detection, schema checks, and merge operations. For UUID mapping of new products, a user-friendly Google Sheet interface was developed. This allowed the client to manually assign UUIDs, and trigger a backend App Script to push those mappings directly into BigQuery, seamlessly feeding the downstream analytics pipeline.