ImeraAI
All case studies

Healthcare · Michigan

A clinic front desk that never misses a call

Clinics lose enquiries to the phone. Aurelius answers the calls and the DMs, fills cancelled slots and brings clients back — so the team can look after the room instead of the handset.

Aurelius AI Front Desk SolutionsMichigan, United States
Visit the live site
A patient checking in with the Aurelius AI assistant at a clinic reception desk
Every enquiry
answered
Every chair
filled

What we did

  • AI voice agent
  • Appointment scheduling
  • Recall and reactivation

Built with

  • Realtime voice pipeline
  • Practice management integration

The challenge

Clinics lose enquiries to the phone. Calls go unanswered while the team is with a patient, DMs sit overnight, cancelled slots stay empty, and nobody chases the client who never rebooked.

What we built

Not a chatbot — a front desk that never misses. Aurelius runs the front of house: it answers the calls and the DMs, fills cancelled slots, confirms visits and brings clients back, so the team can look after the room instead of the phone.

Every clinic runs the same quiet arithmetic. A phone rings while the front desk is checking someone in. It rings again during a procedure. By the end of the day a handful of people wanted an appointment and did not get one — and nobody on the team can name who they were.

What was actually breaking

The failure was never rudeness or effort. It was concurrency. One desk, one person, and three channels of demand arriving at once:

  • Calls during treatment, answered by voicemail nobody returns.
  • Direct messages that land after hours and are read the next afternoon.
  • Cancellations that open a slot at 9am for an appointment at 2pm, with no time to fill it by hand.
  • Lapsed clients who never rebooked, and were never chased because chasing is the first task to fall off a busy day.

What we built

Aurelius sits in front of all four. It answers on the first ring, holds a real conversation about what the caller needs, checks the schedule, and books.

When a slot opens, it works the waitlist rather than waiting for someone to notice. Confirmations and reminders go out on their own. Clients who have drifted get a recall message that reads like the practice, not like a marketing tool.

Anything clinical, sensitive or unusual routes to a human immediately — the agent's job is the front of house, not the consulting room.

Why it held up in a clinic

Two things mattered more than the model:

  1. Latency. A front desk that pauses for two seconds before answering reads as broken. The voice path was budgeted end to end so the first response lands inside the window a caller reads as "someone picked up".
  2. Writing into the real schedule. A booking that lives only in a transcript is not a booking. Every confirmed appointment is written into the practice's own system, where the team already looks.

Where it stands

The front desk now answers regardless of what the room is doing, and the schedule fills without anyone working a callback list at the end of a shift.

More work

Financial Services · In discussion

An AI virtual assistant that an auditor can read

Front-line servicing conversations handled by AI avatars and virtual receptionists, with controlled workflows, human escalation on anything sensitive, and a decision log behind every interaction.

Read

Home Services · California

Answering on the first ring, at 11pm, in California

Home service demand arrives by phone, usually outside office hours, and the first company to answer wins the job. Mango AI answers, qualifies, checks availability and books straight into the schedule.

Read

Let's build something that holds up.

Tell us what breaks in your operation today. If it is the kind of problem we build for, we will say so and if it is not, we will say that too.