A busy schedule hides a lot. A practice can be running full days and still be losing money in two specific places, both of which are invisible unless someone is measuring past the ad account.
The first is new patients that cost more to acquire than they are worth. The second is existing patients quietly falling out of recall. The second is usually the larger number, and it is the one nobody is looking at.
Recall is the whole business
A patient who stays in recall is worth many times a new patient acquisition, and the gap compounds every year they stay. That makes recall the most valuable system in the practice — and in most practices it is handled by a person with a list, which means it happens in quiet weeks and stops entirely in busy ones.
Automating recall does not make it better than a good front desk on a good day. It makes it consistent, which a busy front desk never is. The patient who is due gets contacted whether or not anyone had time, and the front desk is freed for the conversations that need a person.
Reactivation is the same argument applied to patients who have already lapsed. Every practice has a list of people who came regularly and then stopped, usually for no dramatic reason. Working that list systematically produces production at a fraction of the cost of new acquisition.
Cost per new patient, not cost per lead
The advertising report will show you cost per lead. That number is not actionable, because a lead is not a patient. Some leads never book, some book and never attend, and some attend once for an emergency and never return.
What you need is cost per new patient by channel, and for the higher-value work, cost per case. This requires the booking and attendance data to connect back to the marketing source, which is real setup work and is worth doing once properly.
Without it, budget flows to whichever channel produces the cheapest enquiries — which is frequently the one producing the least valuable patients.
High-value cases behave completely differently
Implants, orthodontics, and cosmetic work have long consideration cycles, high values, and a different decision process from a hygiene appointment. Running them in the same campaigns as routine work, judged on the same window, hides both.
They need their own campaigns, their own landing pages, and follow-up built for a decision that takes weeks. A three-day nurture sequence designed for an emergency enquiry is the wrong tool for someone deciding on a five-figure treatment plan.
There is also usually a substantial amount of production sitting idle in the practice software as accepted-but-unscheduled and presented-but-undecided treatment. Sequences aimed at those two groups are among the highest-return automations available, and almost nobody runs them.
Health advertising has rules, and patient data has more
Platform policy on health and personal attributes constrains dental creative in ways that catch practices out. Copy addressing the reader's teeth or appearance directly is a common rejection trigger even when the claim is accurate.
Separately, patient information carries privacy obligations that shape what marketing systems should ever touch. The right design keeps marketing automation running on the minimum information needed and does not move clinical detail into systems that have no business holding it. A marketing partner that shrugs at this is creating a problem you will own.
What to fix first
If the schedule has gaps, the order is almost always:
First, recall automation, because it works on people who already chose you.
Second, reactivation of lapsed patients, for the same reason.
Third, unscheduled treatment follow-up, because that production is already sold and just not booked.
Fourth, no-show reduction — confirmations, reminders, recovery.
Only then, acquisition, and split by treatment type rather than run as one campaign.
Spending on new patients while the recall system leaks is refilling a bucket with a hole in it. The hole is cheaper to fix than the water is to replace.