Financial services provider
A Financial Services Provider: Cutting the Friction Out of Loan Applications
A financial services provider moved loan intake to an AI voice assistant that collects, verifies and routes applications in a single call. Processing time fell by 85 per cent.
Timeline: Six-week deployment roadmap
- 85% Reduction in application processing time
- 5 hours Saved per application
- 98% Data capture completeness in one conversation
- 300% Increase in completed applications
- ~$100K Estimated annual saving (60% lower operational costs)
- 3x Applications processed with the same workforce
The problem
Loan applications were long and frustrating. Teams were overwhelmed with manual verification, legacy systems created bottlenecks, and lengthy paperwork and repeated callbacks led applicants to abandon mid-process. Manual data entry caused errors, compliance issues and costly rework, and document checks took hours per application.
- Manual data entry errors caused compliance issues and rework
- Applicants abandoned due to paperwork and repeated callbacks
- Legacy systems created processing bottlenecks
- Manual document checks took hours per application
What we built
We built a voice-led intake system. Applicants call a dedicated Twilio number and speak with a VAPI AI voice assistant, which collects personal, employment and financial details in a natural conversation. ChatGPT reviews the data for context and eligibility, and n8n verifies the information, updates Airtable and syncs to Salesforce in parallel. n8n then routes each application to one of three queues: instant approval, manual review, or a follow-up for additional information, with a full audit trail for compliance.
Tools: Twilio, VAPI, n8n, ChatGPT, Airtable, Salesforce
How we ran it
Every engagement follows The Claro Build Framework.
- Assess: We examined the application journey and identified manual verification, legacy bottlenecks and repeated callbacks as the main drivers of abandonment and rework.
- Design: We designed the voice conversation, the validation logic and the three-way routing rules for approval, review and follow-up.
- Build: We trained the voice assistant, set up Twilio and the CRM integration, built the validation and workflow automation with testing, then completed compliance verification before launch.
- Sustain: Handover followed The Adoption Standard: a structured 45-minute handover call, with documentation written as the system was built.
The result
Application processing time fell by 85 per cent, from weeks to minutes, saving five hours per application. Data accuracy improved by 90 per cent, with 98 per cent of required data captured in a single conversation. Completed applications rose by 300 per cent, teams can process three times as many applications with the same workforce, and operational costs fell by 60 per cent, an estimated saving of $100,000 a year.
Client name withheld for confidentiality.