Updated 2026-08-26 · 4 min read
It's half past eight, the first patient is already in the chair and the phone rings: the twelve o'clock can't come, is there anything in the afternoon? Somebody says «I'll change it later». That «later» competes for the rest of the day with two emergencies, a treatment plan and a call from the lab, and it often loses.
Opening the practice management software mid-morning isn't realistic in a clinic. Replying to a message is.
Replying in the chat that's already open
If the practice assistant sends alerts on WhatsApp about what's happening — an emergency, a patient asking for a person, the morning summary — that conversation is already open on the principal's phone. What's new is being able to reply there with what you want done: «move my twelve o'clock patient to five», «block Thursday afternoon, I have a conference», «tell Gabriel he can come in at nine tomorrow».
The value isn't the convenience, it's the timing: the order is given when it is thought of, which is the only moment you're sure to remember it.
The five orders that make sense
An assistant doesn't need to do everything by chat. It needs the five you ask for with the practice in full swing: move an appointment, book a new one, cancel it, block off a slot in the diary and message a patient. Everything else — reports, treatment plans, campaigns — is better done sitting in front of the dashboard.
Notice which of those five are reversible and which aren't. Moving, booking and blocking are undone in two clicks. Cancelling and messaging a patient are not. That line is what should govern your team's permissions.
What has to happen BEFORE it acts
This is where the serious tools part company from the frightening ones. An assistant that gets «move the twelve o'clock» and moves it straight away is a risk: you only need two patients with the same name, or a message typed in a hurry, for somebody else's appointment to move.
What it should do is send back what it is about to do, with the specifics: the patient's full name, the day and time before and after, the dentist. If that wasn't what you meant, you see it there and say no. And if two patients match, it must ask which one instead of choosing for itself.
One condition almost nobody weighs up when signing: that confirmation has to EXPIRE. If the message goes unread and «yes» is sent the next morning, the last thing you want is an appointment moving that was already sorted another way. Ten minutes is a sensible window.
Three limits worth insisting on
- It mustn't message any patient in the middle of the night. The principal may be awake at three thinking about the practice; the patient doesn't have to get a WhatsApp at that hour. A good assistant says so and sends it in the morning, instead of quietly queueing it and leaving whoever asked thinking it already went out.
- Each phone should have its own role. The principal's phone and the reception phone shouldn't be able to do the same things: moving, booking and blocking are reversible; cancelling and messaging a patient are not. If the tool doesn't tell phones apart, the whole practice has the highest permission.
- It has to be logged. Every order carried out, with the time and the number it came from — that's what turns «I think I changed it» into something you can check two weeks later.
What this is NOT
It isn't an assistant that runs the practice on its own. The judgement stays with the team: you ask, you confirm and you can say no. What changes is that decisions can be made on your feet instead of piling up for the end of the day. And it has to be stoppable with one sentence: with ClaudIA you write «stop doing what I say» and it stops carrying anything out, while still sending you alerts.
How to try it without risk
Start with the reversible: in the first week, only diary blocks and the odd time change. If something comes out odd, it's undone. Once you've seen that the proposal it returns always says what you meant, move on to cancellations and patient messages. And look at the log at the end of the week — not to be suspicious, but to count how many things got sorted on the spot that would previously have waited until closing. That number decides whether it's worth it for you.