Normalized claim
Tax documents processed per minute: 5,000,000 documents/minute increase
It processes 5 million tax documents per minute
MicrosoftEvidence: Medium50/100
BMTax processes 5 million tax documents per minute with Microsoft Azure OpenAI in Foundry Models
Use case typeData platform modernizationUpdated Jun 29, 2025
BMTax is a Brazilian tax and fiscal consulting firm that was struggling with fragmented IT infrastructure and manual, knowledge-dependent processes amid major tax reforms. The company and Microsoft partner Qualiserve consolidated 45 TB of data into Microsoft Fabric in six months and built Intelligence Tax Hub, an internal platform with an Azure OpenAI in Foundry Models chatbot for natural-language data discovery and reasoning. The platform also uses Power BI and Microsoft security solutions to support analytics, governance, and sensitive-data protection.
Reported outcomes
Productivity: +15–25%
Productivity & throughput
Tax documents processed per minute: 5,000,000 documents/minuteFiscal records analyzed: More than 2,000,000,000,000 BRL
Catalog median for productivity & throughput deployments: +40% across 100 reported metrics. Compare benchmarks →
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Productivity: 15-25% increase
BMTax initially increased productivity by 15–25%
Normalized claim
Manual reconciliation inconsistencies: 20-35% decrease
reducing the number of inconsistencies that employees manually reconcile by 20–35%
Normalized claim
Fiscal records analyzed: 2,000,000,000,000 BRL increase
has analyzed more than R$2 trillion (almost USD400 billion) in fiscal records
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- BMTax
- Provider
- Microsoft
- Maturity
- Unknown
- Linked source
- Microsoft Customer Stories
No explicit deployment-stage evidence found.
Customer identity supportedSource describes one deploymentMaturity evidence evaluated
Primary read
Use case focus
Showing 2 of 2
- 1Data platform modernization
- 2Conversational analytics
- BMTax worked with Microsoft partner Qualiserve to build Intelligence Tax Hub on Microsoft Fabric, unifying 45 TB of data into a centralized lakehouse within six months.
- The platform includes an internal chatbot powered by Azure OpenAI in Foundry Models to help employees discover and interact with data using natural-language retrieval and reasoning.
- Power BI provides analytics and visualization, while Microsoft security solutions help protect sensitive, highly regulated data.
- BMTax says it is also building specialized AI agents on top of the platform for intelligent querying and scenario simulation.
- The platform processes 5 million tax documents per minute.
- BMTax initially increased productivity by 15–25% after removing manual processes.
- Improved traceability reduced the number of inconsistencies employees manually reconcile by 20–35%.
- BMTax has analyzed more than R$2 trillion in fiscal records.
- The firm says the platform creates a foundation for new data- and automation-based services and better tax risk/opportunity identification.
Architecture
Intelligence Tax Hub is built on Microsoft Fabric with 45 TB of data unified in a centralized data lake. Azure OpenAI in Foundry Models powers an internal chatbot that uses natural-language retrieval and reasoning. Power BI supports analytics and visualization, and Microsoft security solutions protect sensitive data. The article also notes plans for specialized AI agents for querying and scenario simulation.
Implementation partners1
Sources & evidence1
Evidence: Medium50/100Evidence strength
- Customer explicitly identified
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 29, 2025Publisher: MicrosoftEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.