Modernize unstructured data workloads with Alteryx and BigQuery

Normalize data
The extracted invoice numbers are cleansed and standardized so that minor formatting differences — such as whitespace or inconsistent formatting — do not hide potential duplicates during reconciliation.

Compare against history
The workflow joins the normalized current invoice records against historical invoice data already resident in BigQuery. This reconciliation step identifies records with matching invoice numbers already present in history and flags them as potential duplicates.

Flag exceptions
A formula step then applies business-rule logic to the remaining records. Those rules can vary by workflow and may include checks for amount mismatches, missing required fields, invalid totals, or other finance-specific exceptions. In this example, one rule checks whether invoice_amount matches net_amount + tax_amount and flags mismatches for review.

Produce operational outputs
The workflow produces two operational outputs: one for invoices matched to historical records and flagged as potential duplicates, and another for unmatched current invoices that have been validated, labeled for exceptions where needed, and prepared for unpaid-invoice processing.

The result is not just extracted data, but governed operational outputs that finance teams can review and act on immediately.

Summary: A governed document-to-decision workflow

Our joint customers are excited for these new capabilities that will help improve business processes. 

“What stood out to me about Alteryx One: Google Edition was how naturally it fits into the Google Cloud experience. I’m excited about the opportunity to make analytics more accessible by enabling business users to work with BigQuery data through an intuitive, governed experience.” – Michael Wyant, Vice President, Enterprise Data & Corporate Solutions, Papa Johns

This use case is more than just a simple extraction tool; it represents a strong Google Cloud-native pattern for transforming unstructured documents into actionable operational outputs. Instead of moving warehouse data into disconnected preparation tools, the workflow keeps storage, extraction, reconciliation, and output generation aligned with governed enterprise data already resident in BigQuery. By combining the intuitive design of Alteryx One Live Query with the scale and AI capabilities of BigQuery, finance teams can reduce manual effort, accelerate exception handling, and focus on higher-value strategic initiatives.