Customer support · Production

Retell AI cuts Sunshine Loans call abandonment from 30% to under 5% and automates 75–80% of inbound support

The problem

Sunshine Loans depended on costly offshore call centers that could not keep up with rising contact volume. Abandonment rates ran as high as 20–30% and the offshore model failed to deliver the customer respect the brand aimed for.

First attempt

Even with 100+ offshore agents, abandonment rates hit 30% and customer satisfaction was inconsistent, while the offshore model proved incompatible with the company's customer-respect values.

Workflow diagram · grounded in source
1
Inbound call arrives
trigger
“inbound voice agent capable of handling the majority of inquiries”
2
AI voice agent resolves call
ai_action
“75–80% of calls are fully resolved by AI, no human needed”
3
Multilingual agent handles non-English calls
ai_action
“Expanded into non-English-speaking markets with multilingual AI voice agents”
4
Business tool integration
integration
“Integrated critical business tools like Salesforce and Make.com for real-time, end-to-end automation”
5
Compliance audit record
output
“It's better for compliance…we can see everything and report it really easily”
6
Route complex work to skilled staff
routing
“This freed up their in-house staff to focus on high-value work while giving customers faster, smarter answers”
Reported outcome

Retell AI now handles 75–80% of inbound calls fully autonomously, cutting abandonment rates to around 5–6%, replacing 100+ offshore agents with 24/7 AI coverage, and enabling multilingual expansion into non-English-speaking markets.

Reported metrics
call abandonment rate before AI20 to 30%
call abandonment rate after AI5 or 6%
calls resolved by AI75–80%
Monthly applications handledmore than 700,000
Show all 7 reported metrics
call abandonment rate before AI20 to 30%
call abandonment rate after AI5 or 6%
calls resolved by AI75–80%
monthly applications handledmore than 700,000
offshore agents replaced by AI100+
prior outsourced phone support costspending millions on outsourced phone support
deployment time with one junior developerunder a month with just one junior developer
Reported stack
Retell AISalesforceMake.com
Source
https://www.retellai.com/case-study/sunshine-loans-retell-ai-customer-support-automation
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Retell AI now handles 75–80% of inbound calls fully autonomously, cutting abandonment rates to around 5–6%, replacing 100+ offshore agents with 24/7 AI coverage, and enabling multilingual expansion into non-English-sp…

What tools did this team use?

Retell AI, Salesforce, Make.com.

What results were reported?

call abandonment rate before AI: 20 to 30%; call abandonment rate after AI: 5 or 6%; calls resolved by AI: 75–80%; Monthly applications handled: more than 700,000 (source-reported, not independently verified).

What failed first in this deployment?

Even with 100+ offshore agents, abandonment rates hit 30% and customer satisfaction was inconsistent, while the offshore model proved incompatible with the company's customer-respect values.

How is this customer support AI workflow structured?

Inbound call arrives → AI voice agent resolves call → Multilingual agent handles non-English calls → Business tool integration → Compliance audit record → Route complex work to skilled staff.

Related customer support cases