Kyc aml · Production

Banorte cuts document validation time 60% with SS&C Blue Prism AI, ML and NLP automation

The problem

Banorte employees manually validated documents for payroll loan and credit card requests at high volume, with each analyst receiving up to 100 new requests per 1–1.5 hours, making it difficult to meet SLAs and respond to customers promptly.

Workflow diagram · grounded in source
1
Employee submits request notification
Trigger
When an employee receives a request for a new payroll loan, instant-use credit card, or payroll improvement, they notify the SS&C Blue Prism AI agent.
▾ source quote
“when an employee receives a request for a new payroll loan, an instant-use credit card or payroll improvement, they notify an SS&C Blue Prism AI agent”
2
AI agent validates customer data
Validation
The SS&C Blue Prism AI agent validates the customer data via an internal banking application.
▾ source quote
“who validates the customer data via an internal banking application”
3
Documentation downloaded and uploaded
Integration
The AI agent downloads key documentation such as official identification and proof of address and uploads it to Banorte's internal AI engine.
▾ source quote
“the AI agent downloads key documentation — like official identification and proof of address — and uploads it to Banorte's internal AI engine”
4
AI engine processes documents
Ai action
Banorte's internal AI engine identifies the documents, interprets text and images, and extracts the required data.
▾ source quote
“The AI engine identifies the documents, interprets text and images, extracts the required data and sends it back to the SS&C Blue Prism AI agent”
5
Multi-system data aggregation
Integration
The AI agent combines the extracted document data with information from three other systems.
▾ source quote
“The AI agent uses this data, along with information from three other systems”
6
Checklist creation and approval
Output
The AI agent creates a checklist and approves the customer's request.
▾ source quote
“to create a checklist and approve the customer's request”
Reported outcome

Document validation time dropped 60% from 20 minutes to eight minutes per request, returning 2,000 hours per month to the business.
Processing capacity grew 30% year over year without additional headcount, and automatic approval rates rose from 70% in 2022 to 91% in 2024.

Reported metrics
Document validation time reduction60%
Document validation time before automation20 minutes
Document validation time after automationeight minutes
Hours returned to business per month2,000 hours (250 days)
Show all 9 reported metrics
document validation time reduction60%
document validation time before automation20 minutes
document validation time after automationeight minutes
hours returned to business per month2,000 hours (250 days)
processing capacity increase year over year30%
automatic approval rate 202270%
automatic approval rate 202491%
analyst manual request volume (pre-automation)100 new requests in 1 to 1.5 hours
market adaptation turnaround timeless than 24 hours
Reported stack
SS&C Blue PrismAIMLNLPMicrosoft Azure
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://www.blueprism.com/resources/case-studies/banorte-automation-ai-document-processing/
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Document validation time dropped 60% from 20 minutes to eight minutes per request, returning 2,000 hours per month to the business.

What tools did this team use?

SS&C Blue Prism, AI, ML, NLP, Microsoft Azure.

What results were reported?

Document validation time reduction: 60%; Document validation time before automation: 20 minutes; Document validation time after automation: eight minutes; Hours returned to business per month: 2,000 hours (250 days) (source-reported, not independently verified).

How is this kyc aml AI workflow structured?

Employee submits request notification → AI agent validates customer data → Documentation downloaded and uploaded → AI engine processes documents → Multi-system data aggregation → Checklist creation and approval.

Related kyc aml cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Kyc aml cases →