Customer support · Production

Cents builds 24/7 AI voice receptionist 'Assist' on Retell AI for 4,000+ laundromat operators

The problem

Laundromat operators receive constant customer calls around the clock but cannot afford full-time front-desk staff to answer them, leaving questions unanswered and orders lost to competitors who pick up the phone.

First attempt

Cents attempted to build an in-house voice solution by combining LLMs, speech models, and telephony logic, but found the complexity, latency challenges, and maintainability prohibitive and abandoned the effort.

Workflow diagram · grounded in source
1
Customer calls laundromat
trigger
“Customers call around the clock asking if the store is open, where it's located, or whether pick-up service is available.”
2
AI voice agent answers call
ai_action
“Cents launched Assist, their AI-powered voice receptionist built on top of Retell AI”
3
Immediate response delivered
output
“End customers also welcomed the immediate response and often commented on how natural and human the agent sounded.”
Reported outcome

Assist became one of the most popular features in Cents' product suite, giving operators breathing room and acting as a strong growth engine through significantly improved conversions after trade show and demo campaigns.

Reported metrics
calls resolvable by AI without a humanmore than 75 percent
Conversion improvement from demo environmentsignificantly improving conversions
time to launch Assistfour weeks
Operator daily workloadbreathing room in their day-to-day operations
Reported stack
Retell AILLMs
Source
https://www.retellai.com/case-study/how-cents-transforms-laundry-operations-with-24-7-voice-automation
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Assist became one of the most popular features in Cents' product suite, giving operators breathing room and acting as a strong growth engine through significantly improved conversions after trade show and demo campaigns.

What tools did this team use?

Retell AI, LLMs.

What results were reported?

calls resolvable by AI without a human: more than 75 percent; Conversion improvement from demo environment: significantly improving conversions; time to launch Assist: four weeks; Operator daily workload: breathing room in their day-to-day operations (source-reported, not independently verified).

What failed first in this deployment?

Cents attempted to build an in-house voice solution by combining LLMs, speech models, and telephony logic, but found the complexity, latency challenges, and maintainability prohibitive and abandoned the effort.

How is this customer support AI workflow structured?

Customer calls laundromat → AI voice agent answers call → Immediate response delivered.

Related customer support cases