A front desk that never puts a patient on hold.

The calls you are missing

A patient calling to book is rarely loyal yet. Someone who has just been told by their doctor to see a physiotherapist, or who woke up unable to turn their head, is calling clinics in order until one of them answers. If your line is busy and the voicemail box is full, the appointment goes to the clinic down the road, and you never learn it happened.

The second cost is quieter. Every call your front desk answers while a patient is standing at the counter is a small hit to the experience of the person in front of them, and to the accuracy of whatever they were in the middle of entering. A busy desk is not a sign of a healthy clinic; it is usually a sign that one person is doing three jobs at once.

Both problems have the same shape: calls arriving when there is no capacity to take them. That is a scheduling problem before it is a technology problem, and it is worth being honest about which one you have. If your desk is fully staffed and your phones are quiet, an AI receptionist will not create demand that does not exist.

What it actually handles

An AI receptionist answers the phone in a natural voice, identifies itself as an automated assistant, and works through the calls that follow a predictable script. In a clinic, that is most of them.

It also fills the gap after hours, which is often the single largest source of value. An evening or weekend caller who can book a Tuesday appointment without waiting for Monday morning is an appointment you would otherwise have had a coin-flip chance of keeping.

Where it earns its keep a second time is the outbound side: reminders before appointments, and calling or texting a waitlist when a cancellation opens a slot. A gap filled the same day is revenue that was already on the schedule and would otherwise be lost.

What stays human

The line we hold is simple: an AI receptionist handles scheduling and logistics. It does not practise.

• Anything clinical. Whether a symptom needs urgent attention, whether a treatment is appropriate, how an injury should be managed — these go to a practitioner, every time, with no attempt at a helpful guess.
• Emergencies. The system needs a clear, tested path that recognises an urgent call and directs the person to appropriate care immediately rather than booking them for Thursday.
• Complaints and distress. A frustrated or upset patient should reach a person quickly. Automated cheerfulness makes those calls worse.
• Anything the caller wants a human for. If someone asks for a person, they get one — that is a setting, not a debate, and it should be on by default.

A patient should also always be told they are speaking with an automated assistant, at the start of the call. Clinics that hide it get one round of novelty and a long tail of irritation.

Privacy, in Canada, in plain terms

Health information is the most sensitive category of personal information a small business handles, and clinic phone calls are full of it. In British Columbia, private-sector clinics generally fall under provincial privacy legislation — BC's Personal Information Protection Act — with PIPEDA in the picture for federal and cross-border considerations. Your college may also set its own expectations for patient records and communications.

We are not your privacy lawyer and this page is not legal advice. But there are five questions we insist on answering before a clinic system goes live, and you should ask any vendor the same ones.

• Where does the data live? Which country the call audio, transcripts and any patient details are processed and stored in, and whether that can be kept in Canada.
• How long is it kept? Recording and transcript retention should be a deliberate choice with a set period, not a vendor default of forever.
• Is it used to train models? Get the answer in writing. For patient calls, the answer needs to be no.
• Who can see it? Which of your staff, which of the vendor's, and what happens to access when someone leaves.
• What does the caller get told? The disclosure at the start of the call, and what happens if they say they would rather not be recorded.

None of this is exotic, and none of it slows a project down much. It is simply the part that a demo will not cover, and the part you cannot retrofit after a year of stored call recordings.

Your booking software is the deciding factor

Most clinics in this region run on Jane, and a good number run on something else entirely. This is the detail that decides whether an AI receptionist is genuinely useful or quietly creates work, so it is worth getting specific before you sign anything.

The question to ask is not "do you integrate?" Almost every vendor says yes. The question is what happens at the moment of booking: does the system write the appointment into your schedule directly, does it send the patient a link and let them book themselves, or does it hand a message to your front desk to enter manually? All three can be reasonable. Only the first one saves your team the work, and only some software supports it.

We check that against your actual setup before recommending anything, including the awkward parts — practitioner-specific availability, initial assessments versus follow-ups needing different appointment lengths, and whether new patients are allowed to self-book at all. Where a direct write-in is not available, a booking-link handoff is usually still worth doing; it just needs to be described accurately rather than sold as full automation.

By specialty

Clinics are not interchangeable. What callers ask about, how insurance works, how a course of care is scheduled and what a no-show costs all differ enough to change how the system should be built.

Why only one specialty page is live. Physiotherapy is the one where we already have real, specific research and live work. The rest get published as each earns its own material — genuinely different numbers, examples and answers, not a template with the specialty name swapped in.

Getting started

Every clinic build is scoped as a defined piece of work, agreed in writing before anything begins, and shaped by what your call volume and your practice-management software actually allow. That is a conversation about your schedule, not a package chosen off a page.

Ongoing tuning matters more here than in most automation work, because the questions patients ask change with the seasons and with whatever your clinic starts offering. That can run as an ongoing arrangement with us or move in-house once the system is stable — whichever suits how your front desk works.