An average dental clinic gets between fifteen and thirty WhatsApp messages a day asking for an appointment. Almost all of them land on the same phone, and that phone belongs to the front desk — the same person looking after the patients already sitting in the waiting room.
The result won’t surprise anyone who’s worked in one: morning messages get answered mid-afternoon, evening ones the next day, and the patients who messaged three clinics already booked with whoever replied first.
The question isn’t whether that can be automated, because it can. The question is which parts are worth handing off and which ones you shouldn’t touch.
The line between the two
You automate what needs consistency. You don’t automate what needs judgement. Put like that it sounds obvious, but the line gets crossed easily in practice, because the things that need judgement sometimes arrive disguised as simple questions.
On the automatable side sits everything repetitive: hours, address, payment methods, whether you take a particular insurance plan, whether there’s parking, how long a standard procedure takes. In most clinics that’s between 40% and 60% of message volume. Also on that side: checking real availability against the calendar, booking end to end, sending the reminder twenty-four hours ahead — the single biggest lever on no-shows and the first thing dropped when things get busy — and making rescheduling frictionless, because a patient who tells you in advance frees a slot you can refill.
On the other side sits anything clinical. Any question about diagnosis, current pain, or medication goes to a person with no exceptions, and urgent cases too: someone in acute pain needs to speak to a human now, and the agent has to recognise that and prioritise it rather than processing it as one more booking. Quotes for complex treatments need an assessment, so the agent explains the process and books the consultation but doesn’t quote a final price. And with the patient who’s been coming for years, it’s better that someone from the team replies, because that relationship is part of why they keep coming back.
What the flow looks like when it works
A message arrives at whatever hour, Sunday at eleven at night included. The agent works out what that person wants and tells apart someone booking, someone asking a price, and someone reporting pain, escalating the third immediately.
If it’s a booking, it checks the clinic’s actual calendar — with that day’s blocks and openings, not a copy or a generic schedule — offers concrete times, takes the details, blocks the slot, and sends the confirmation. A day before, it sends the reminder with a one-tap reschedule option.
When the front desk arrives in the morning, the day has already started: last night’s messages are answered and there are appointments on the calendar.
What you need in place first
This part gets skipped often and causes the most trouble later.
You need a digital calendar, which can be Google Calendar, dental practice software, or even a well-structured spreadsheet; what doesn’t work is a paper diary. You need clear booking rules — how long each appointment type runs, which slots stay reserved for emergencies, how many patients per practitioner — and if those rules only live in the front desk person’s head, they need writing down before you automate anything. And you need a WhatsApp Business line, ideally separate from anyone’s personal phone.
Above all, you need to have decided what gets escalated and to whom. It’s the most important decision in the project and the one most often postponed, usually because it forces uncomfortable conversations about who answers what.
Measure first, or you won’t know whether it worked
This is the step almost nobody takes, and without it any later result is an impression rather than a fact.
For two weeks, before changing anything, write down how many messages come in per day, your average time to first reply, how many arrive outside working hours, how many conversations end with the patient’s message unanswered, how many of every ten enquiries turn into a booking, and how many booked appointments were no-shows. Six numbers, all available from your chat history and your calendar, and if in three months you want to know whether this was worth it, they’re the only thing that will let you answer.
There’s a fast and fairly revealing version of the same exercise: ask someone who doesn’t work at the clinic to message as a new patient on a Tuesday at three in the afternoon and note how long the reply takes. It’s the cheapest diagnostic you can run and usually the most uncomfortable.
When it isn’t worth it
Worth saying as plainly as everything else. If your clinic gets fewer than five booking enquiries a day, the front desk is probably coping and the money goes further attracting patients than handling them. If the problem is that patients aren’t coming, automating bookings doesn’t fix it — it makes a funnel that’s empty at the top more efficient. And if nobody’s clear on how long each procedure takes or how the calendar is structured, sort that out first, because automating on top of a confused process just produces badly booked appointments faster.
About the front desk
The most common worry we hear is whether this replaces that person, and in practice the opposite happens. We don’t say that to be diplomatic: what leaves their day is answering “where are you located” forty times, not the part of the job that needs a human.
They’re still the one handling judgement calls, greeting patients, and resolving anything complicated. The difference is they stop being tied to a phone while someone waits at the counter.
At Katalis we build booking agents for clinics, connected to WhatsApp and the calendar you already use. If you’d like to review your current process before deciding anything, get in touch — the first call is a diagnostic, not a sales pitch.