What challenges did Pernambucanas face before AI implementation?

Automate interpretation and verification of forged or falsified ID documents for customer admissions. Eliminate manual 24/7 document checking while reducing costs, latency, and workload. Improve the quality of collected data used for fraud detection.

Source

GCPProductionEvidence: Medium65/100

Pernambucanas (Pefisa): Documentoscopia AI fraud detection using OCR + Vision AI / Document AI

Use case typeDocument automationUpdated Jun 13, 2026

Pernambucanas is a Brazilian retailer and fintech operator whose Digital Lab built and deployed Documentoscopia, a cloud-based tool to validate ID documents for customer onboarding and fraud prevention. The solution integrates Google Cloud OCR capabilities and Vision AI with Pernambucanas's own applications to recognize and extract text from printed or scanned documents, then uses that extracted data as input to fraud detection models. Documentoscopia runs 24/7 and was built to replace fully manual document validation, improve scalability, reduce costs and latency, and raise data quality during customer admission.

Organization: Pernambucanas
Industry: Retail
Location: Brazil
Published: June 2025

Reported outcomes

80%

quantified impactAutomation & deflection

10 secondstime

Strategic outcomes

Risk & complianceAutomated forged document verificationCost efficiencyReduced manual document checkingNew product / capabilityBuilt in-house fraud detection toolScale & capacityEnabled round-the-clock document processing

Catalog median for automation & deflection deployments: +49% across 19 reported metrics. Compare benchmarks →

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Quantified impact: 80%

Google Cloud Customer StoryJun 2, 2025Customer storyInferred claimMedium evidence strength

Specialized manual effort to identify document-based fraud fell by 80%.

Normalized claim

Time: 10 seconds decrease

Google Cloud Customer StoryJun 2, 2025Customer storyInferred claimMedium evidence strength

Average customer acceptance time was reduced from hours to up to 10 seconds.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Pernambucanas, Pefisa
Provider: GCP
Maturity: Production
Linked source: Google Cloud Customer Story

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Document processing
2Fraud detection
3Customer onboarding

Automate interpretation and verification of forged or falsified ID documents for customer admissions.
Eliminate manual 24/7 document checking while reducing costs, latency, and workload.
Improve the quality of collected data used for fraud detection.

Pernambucanas's Digital Lab built Documentoscopia in-house.
The team integrated Google Cloud OCR capabilities with its own applications to extract text from printed and scanned IDs.
The extracted fields feed fraud detection models, and Vision AI is used to improve image analysis and document processing.
The workflow operates 24/7 in a fully cloud-based setup.

Technologies

Google Cloud OCR Vision API Documentoscopia Apigee API Platform

Document analysis time dropped from a few days to an average of about eight seconds.
The company processes more than 10K documents per day, about three documents every two seconds.
Specialized manual effort to identify document-based fraud fell by 80%.
The OCR solution has processed over 14 million elements and about 300K document images per month.
Average customer acceptance time was reduced from hours to up to 10 seconds.

Architecture

Pernambucanas's Digital Lab built Documentoscopia and integrated Google Cloud OCR capabilities with its own applications. The extracted document text is used as input to fraud detection models, and Vision AI supports image analysis in a 24/7 cloud-based onboarding workflow.

Sources & evidence1

Evidence: Medium65/100Evidence strength