Clinic owners talk about cost per lead. The number that matters is cost per accepted treatment plan, and between the two sits a path that most practices have never measured: the enquiry form, the reply, the consultation booking, the coordinator's close. This article puts the published acquisition numbers side by side and shows where the loss usually is.

| Specialty | Cost per lead (search ads) | Blended cost per new patient | Source |
|---|---|---|---|
| Plastic surgery | $102.51 | ~$610; consultation-level quotes to $1,200 | LocaliQ 2024–25; Dentplicity Jan 2026; Cakesmash Jun 2026 |
| General dentistry | $84.77 ($7.03 CPC) | $150–300 (referrals $25–75, SEO $50–150, Google Ads $150–350, social $200–400) | LocaliQ; Dentplicity |
| Orthodontics | $8.76 CPC, highest tracked | n/a | LocaliQ |
| Dermatology | best converter, 25.3% CVR | $250–500 | LocaliQ; ThinkBasis 2025 |
| Medical aesthetics | varies; POM rules limit ads in UK/EU | marketing ~7% of revenue, ~$98K per location per year | AmSpa 2024 |
Two things stand out. Acquisition is expensive everywhere in the category, and retention is cheap by comparison: $35 to $85 to keep a patient against $150 to $600 to find one. A clinic that recalls its existing patients properly is buying growth at a fifth of the price. The second is that the blended figures are several times the cost per lead, which means most of the cost is incurred after the click.
In the practices we audit, the median time to a human reply to a web enquiry is measured in hours, and out-of-hours enquiries often wait until the next working day. Seventy percent of health questions put to ChatGPT arrive outside clinic hours (OpenAI), and a patient who has just asked an assistant which clinic to trust is contacting more than one. The clinic that replies within the hour is usually the clinic that gets the consultation. A response-time standard costs almost nothing and is the highest-payback change in most diagnostics we run.
Enquiry forms that ask nine questions before a phone number lose most of their starters. The practical test is to enquire as a patient on a phone and count the steps. Fewer fields, a visible phone number, a clear statement of what happens next, and online consultation booking where the practice can support it.
Treatment coordinators who quote a price for a clinical procedure by phone lose the consultation to the cheapest quote in the city. The method we teach is strict: listen first, reflect the concern in the patient's words, explain what a consultation involves, book it. Price is discussed when the clinician has set the plan.
Three forces. Auction prices in healthcare search rise every year as more clinics bid on the same procedure terms. Platform rules tighten: Google forbids personalised and remarketing targeting around cosmetic procedures and injections, and Meta restricts lower-funnel health pixel events, so paid media gets blunter as it gets dearer. And discovery is moving into AI answers that name three to five clinics and cannot be bought, so a growing share of patients form a shortlist before any ad is seen.
Take last quarter's marketing spend, agency fees included. Divide by consultations booked, then by treatment plans accepted. Most owners have never seen the second number and are surprised by it. Then ask where along the path the enquiries went. If half of enquiries never reach a consultation, the practice does not have an advertising problem, and buying more enquiries is the most expensive way to find out.
One page per offer: what it is, who it is for, what it costs or how pricing works, and what happens next.
MeasurementTracking from first touch to paid sale, without customer data leaving your systems, with source tagged on every sale.
The partnershipWhere you lose customers between first search and sale, why, and what to do first.