How a multi-location DSO tripled new patient conversion in three months

The group's patient base was largely working parents, who tended to call to book appointments only after the workday ended, picking up the kids, and getting home. By the time most parents had a moment to dial, the call center was closed. The phones kept ringing into the evening, on weekends, and overnight, with no one on the other end.

The team had tried to close the gap by bringing in offshore call center agents to cover the 5pm to 8am window and emergency overflow. The arrangement underperformed from the start. Agents needed continuous retraining, and any gap in that training translated directly into lost bookings and customer service complaints. Agents got sick or went on vacation, leaving the same coverage holes the offshore model was meant to fix. Sourcing Spanish-speaking agents was also a challenge, since many offshore call centers rarely had them.

The result was that the operational problem persisted in a new form. After-hours calls were being answered, but the experience wasn't consistent, and the team was spending significant time managing a workforce that was not always getting better month over month.

The group partnered with Flossy to deploy Fiona, Flossy's AI voice agent, across every location. Fiona handles inbound calls and books appointments directly into the practice's PMS, including after hours, on weekends, and during peak volume when the centralized call center is already at capacity. Fiona is also multilingual by default, handling Spanish, Japanese, and Hindi alongside English, which removed the recruiting problem the offshore model had never fully solved.

The initial deployment focused on the after-hours window, the original pain point, with Fiona picking up calls between 5pm and 8am and on weekends. From there, scope expanded. As the team observed Fiona's performance on after-hours bookings, the deployment was widened to cover daytime overflow and full-call-center coverage across every location.

Unlike the offshore agents, Fiona did not require retraining cycles, sick coverage, or vacation backfill. The Flossy team continuously tuned the system based on real call data, testing which voice patients engaged with, which greetings kept parents on the line, and how to phrase scheduling questions so more parents said yes. The improvements compounded month over month, with the system getting measurably better at every category of call.