How SafeRide Health scaled to a million calls a month with PolyAI
SafeRide Health is a technology and services company that provides non-emergency medical transportation (NEMT) for Medicare and Medicaid health plan members. As the bridge between healthcare payers, providers, and transportation companies, SafeRide Health ensures patients can attend life-sustaining appointments across all 50 states and with more than 40 of the largest health insurers in the country.
Challenge
Sustaining 40% annual growth meant continuously hiring and training CSRs, while January demand spikes pushed call center capacity to its limits.
Solution
An AI agent handling the full range of ride management calls, from authentication to booking and cancellation, scaling instantly to meet demand without adding headcount.
Result
1 million calls handled per month, with speak-to-human turn rate dropping from 55% to 2-3% and 47,000 agent hours saved on authentication alone.
The challenge
Scaling the call center to support 40% growth
SafeRide Health was built on the core belief that technology could do better than outdated manual processes at supporting at-risk health plan members with reliable transportation to dialysis, oncology appointments, and other life-sustaining care.
What started small quickly grew exponentially, with ride and call volume doubling year over year. SafeRide Health built out a call center from scratch to serve its ever-increasing patient base, increasing to over 1,000 customer support representatives (CSRs) across onshore and offshore partners. This sustained 40% annual growth required continuous investment in hiring, training, quality, and cost management, especially as peak demands became more pronounced.
As health plans across the country reset their benefits each year on January 1st, SafeRide typically experiences an influx of call volume nearly overnight. With new-year demand spikes and CSR capacity constraints, SafeRide needed to explore additional technology-based opportunities to meet the demand.
It's very challenging to scale humans in a call center with a 40% a year growth rate. That's a lot of new CSRs to hire, with associated training costs and quality requirements. That's why we had to look for a new way.
The solution
Finding the right AI partner
SafeRide Health needed to find an AI partner with a quality voice, strong engineering collaboration, and a willingness to build a customized solution on SafeRide's own APIs. That’s why they chose PolyAI.
The AI agent, named Steven, was built to handle the full range of ride management calls, including member authentication, booking, cancellation, modification, and flex ride activation. The first version of Steven was designed to be intent-based. The intent-based model worked well, but struggled with the natural speech of patients.
To stay ahead of the curve, SafeRide and PolyAI partnered to make a fundamental change to Steven, upgrading it to be powered by agentic AI. This change enabled Steven to understand the full context of what a member was saying and respond accordingly.
SafeRide Health uses PolyAI’s Agent Studio to manage and optimize the agent without additional engineering needs. Call flow changes and business rule updates that once took weeks are now resolved in seconds. Smart Analyst , the platform's analytics tool, gives the team the same speed on insights, surfaces trends, and finds specific calls through natural language prompts.
If you want a 99% on-time fulfillment rate, a 4.8-star member rating, and 97% on-time performance, you need AI automation. It's the only way you maintain member experience and costs.
The results
Handling a million calls a month
Now three years into the partnership, the impact is visible across every part of SafeRide Health's operations. There has been a 15% increase in success rates across all call flows and a 20–30 second reduction in average handling time. PolyAI also supports 80% of all inbound calls and has helped the business scale without sacrificing the member experience that sits at the heart of everything SafeRide Health does.
Together, SafeRide Health and PolyAI have achieved:
calls answered every month
agent hours saved on authentication
reduction in speak-to-human turn rate
Rather than replacing CSRs, PolyAI has freed them to focus on the calls that genuinely need a human. The complex, high-touch cases, like a mother arranging airfare, hotel accommodation, and meals so her daughter can get cancer treatment, now get the time and attention they deserve.
A member called in and said, 'Sorry to bother you. I usually talk to Steven, but he didn't have my phone number, so I wanted to give it to you so I can talk to Steven next time.’ That's a member who prefers the AI agent, that tells you everything.
Continuing to improve the PolyAI experience
For SafeRide Health, the next chapter is about reaching two key groups to extend AI-powered support to:
- Members who prefer other languages, such as Spanish
- Members with complex benefit rules
Historically, these interactions have required support from a live CSR. Through this partnership, SafeRide is expanding access so these members receive faster, clearer assistance like any other caller.
Beyond those two immediate priorities, the team is exploring how agentic AI can continue to handle increasingly complex call types, while raising the level on what automation can do and the member experience. By making PolyAI the first point of contact for 95% of calls, SafeRide Health focuses on getting members help right away while ensuring a smooth handoff to a human CSR when a human touch is desired or required.
PolyAI has been a fantastic partner on our journey. The fact that we can handle a million calls a month, reliably, with a great experience for our members is a real key differentiator for us, and it allows us to continue to grow.
More customer stories
Start the conversation.
See why companies trust PolyAI to handle their most important conversations.