An agent that works the overdue recall list by WhatsApp and SMS, books into real hygiene openings, confirms tomorrow’s appointments, backfills a cancelled chair from the ASAP list, follows up diagnosed-but-unscheduled treatment, and hands anything clinical or financial to your team. Follow-up only. The dentist keeps every clinical and treatment-plan conversation.
Thirty minutes, free. You leave knowing which workflow to start with, what a pilot covers and what it costs.
PilotAI Agent Pilot: one workflow, 2–3 weeks, $4,500 (₹3,50,000) fixed. What the pilot includes.
Who this is for: General and multi-dentist practices, from a single-chair owner-dentist to a small group, running on practice management software (Open Dental, Dentrix, Curve, CareStack, Practo) with a recall list nobody has time to work.
Patients due for hygiene six to eighteen months ago. Each one is a cleaning, radiographs and the restorative work that visit would have found. The front desk works it on a quiet Tuesday, which never comes.
The ASAP list exists in the software. Working it means six phone calls while two patients wait at the counter. The hygienist reads a magazine and the practice eats the hour.
Treatment plans sit in the system for months. Following up means a conversation about money and anxiety the front desk is not set up to have between phone calls.
Trigger: Monday 7:00 am: 612 patients overdue for recall; 9:40 am: a hygiene cancellation for 11:00.
| Measure | Today | With the agent |
|---|---|---|
| Recall patients contacted per week | ~15, on a good week | The whole overdue list, in batches, with follow-ups |
| Front-desk hours on recall calls, per month | ~20 hours for ~60 attempts | ~3 hours reviewing the log and exceptions |
| Cancelled chairs refilled | Rarely; nobody has the 20 minutes | Attempted every time, within minutes |
| Reminder and confirmation the day before | When there was time | Every appointment, with replies handled |
Worked example. The assumptions:
Front-desk hours are the small number. Production from recovered recall visits, refilled chairs and scheduled treatment is the number that matters, and vendor claims about no-show reductions vary so widely that we measure yours instead: no-show rate, recall rebooking rate and refilled chairs against the previous quarter.
Put your own numbers in the automation savings calculator, or get a price range from the AI agent cost estimator.
Each agent has a small set of tools it is allowed to call and nothing else. That list is the guardrail, and it is written down here so you can see it.
Runs weekly. Pulls the overdue list with the exclusions the practice set, sends personal two-way messages on WhatsApp or SMS, books into real hygiene openings, follows up twice at most, and writes every outcome to the patient record.
Confirms the day before, handles reschedules within the rules, and the moment a cancellation lands it works the ASAP list for that exact slot, books the first yes and closes the loop with everyone else.
Reads treatment plans with no booking, sends a plain, non-pressuring note with a booking link, answers logistics questions, and turns any cost or "do I really need this" question into a task for the treatment coordinator.
We build clinic software for a living: PHMS is our offline-first clinic management system with patient records, appointment queues, prescriptions and a doctor–patient chat. The booking agent on the demos page is the same design as the recall and confirmation agents here. No dental practice has published results with us yet, and this page says so rather than quoting a vendor’s no-show percentage.
They mind being messaged badly: generic blasts, wrong names, messages at 11 pm, no way to reply. The agent sends a personal, two-way message in the practice’s voice, inside quiet hours, stops after two follow-ups, and a reply reaches a person whenever it should. The practice decides whether to say an assistant is answering; we recommend saying so.
Open Dental and Curve have APIs; Dentrix, CareStack and Practo vary by plan and region; many practices run on a mix of software and a calendar. For a pilot the agent can work from the recall report and the ASAP list as exports, with a shared sheet for outcomes, and integrate with the practice system once it has proved itself.
The agent sees the name, the contact preference and the appointment, not the clinical record. Messages stay in your WhatsApp Business account and your practice system; the model provider does not train on your data under the API terms we use. We write the data flow into the handover for your compliance advisor. Where a jurisdiction needs a business associate agreement or consent language, that is set up before go-live.
The AI Agent Pilot puts the recall and confirmation agent on your WhatsApp number in two to three weeks for a fixed fee, measured against your last quarter. Chair refill and treatment follow-up are scoped after. The cost estimator gives you a range in two minutes.
We do not know, and neither do they: the figures in vendor blogs range from 29% to 60% and rarely cite a study. What we do is measure your no-show rate, recall rebooking rate and refilled chairs in the pilot against the previous quarter, and tell you the number. If confirmations with a reply channel do not move it, you will know in three weeks for a fixed fee.
How many chairs, which practice software, and roughly how long the overdue recall list is. We reply with what a pilot on recall and confirmations would cover.
Or start with a number: the AI agent readiness quiz takes three minutes and gives you a score and a 30-day plan.