Dental lead generation: we produced 236 leads at $11.79 and it still failed
Search for dental lead generation and you will find a great many agencies showing you their best month. You will not find many showing you a campaign that worked on paper and failed in the building.
Here is one of ours.
The campaign
A Southern California practice, aligner treatment, paid social. Over the run it produced 236 enquiries at $11.79 each.
By the standard the industry reports on, that is a good campaign and an unarguable one. The cost per lead is low for high-ticket dental. The volume is real. Put those two numbers on a slide with a logo underneath and nobody in the room asks a follow-up question.
Almost none of those people ended up in a chair.
Not "fewer than we hoped". The number of patients who actually attended was so small that the whole exercise was, in business terms, a waste of the practice's money and our time — and we had every ad-side metric telling us it was going well while it happened.
What we got wrong
Three things, and only one of them is about advertising.
We optimised for the cheap lead. A low cost per lead is the easiest number in marketing to improve and the least connected to revenue. Make the form shorter, ask for less, broaden the audience, and the price per enquiry falls while the quality of the enquiry falls faster. We hit $11.79 partly by being good and partly by removing the friction that would have told us who was serious.
We handed the leads to a human and assumed. A person was responsible for calling them. People have other work, and lunch, and Fridays. Enquiries that arrived in the evening were called the following day, or the day after, and by then the enquiry had gone cold in the way enquiries do. This was the real failure and it had nothing to do with the ads at all.
We reported on the wrong thing, so we could not see it. Our own reporting was lead counts and cost per lead, because that is what we could measure easily. Nobody was counting arrivals. The campaign looked healthy for its entire run in exactly the way a system with a dead half always looks healthy — the half that reports is the half that works.
Why cost per lead is the wrong number
A lead is a row in a spreadsheet. A patient is someone in the chair.
The distance between those two things contains: whether anyone called, how fast, how many times, whether they got through, whether a time was agreed, whether anything happened between agreeing and the appointment, and whether the person actually arrived. Every one of those is a place the money leaks, and none of them appears in a cost-per-lead figure.
This is why the useful measure is cost per patient who actually arrived, and eventually cost per accepted case. It is a harder number to produce and an uncomfortable one to publish, which is roughly why it is not the industry standard.
The arithmetic changes shape completely when you use it. A campaign at three times the cost per lead that gets people into chairs beats a cheap one that does not, and it is not close. Cheap leads are only cheap if they arrive.
What actually broke: the response window
The thing that killed that campaign was the hours between an enquiry arriving and anyone speaking to it.
The lead-response research is direct about this. The MIT (Oldroyd) / InsideSales Lead Response Management study found that a contact attempt at five minutes is 21x more likely to qualify an enquiry than one at thirty minutes, that 78% of buyers go with whoever responds first, and that 58% of enquiries are never contacted at all — the average first response taking 47 hours.
Sources: The study · Harvard Business Review
Read our campaign against that and it stops being mysterious. We were not unlucky and the creative was not the issue. We had built a machine for generating intent and no machine at all for catching it, and the intent evaporated on schedule, exactly as that research says it does.
What we changed
Response is a system, not a person. An enquiry is answered inside sixty seconds, at whatever hour it arrives, because that is when high-ticket dental enquiries actually arrive — evenings and weekends, not Tuesday mornings. No front-desk team should be asked to cover that and none can.
Qualifying questions went back in. Slightly fewer enquiries, at a slightly higher cost each, and a much better idea of who is serious. We stopped treating a falling cost per lead as good news on its own.
The booking is protected after it is made. Booked is not arrived. A confirmation cadence runs between the two. At a West Hills, CA practice running one before every appointment, 55 of 91 booked patients arrived — and the difference between that practice and the campaign described above is not the advertising.
We report arrivals. If a number reaches a practice from us, it is a number about people who turned up, and we can walk through how it was produced line by line.
What to ask an agency
Whether it is us or anybody else, four questions separate the ones who own the outcome from the ones who hand over a list:
- Who makes the first contact with an enquiry, and how long after it arrives? If the answer is the practice's front desk, you are buying leads, not patients.
- What is the show-up rate on appointments you generate? If they do not measure it, they are not responsible for it.
- What is your cost per patient who attended? Not per lead. If they cannot produce it, ask what they would need to.
- Can you show me a campaign that did not work? This one tells you the most. Everybody has several.
We publish this one because we would rather be the agency that answers question four than the one that changes the subject — and because the failure taught us the thing our whole service is now built around. The leads were never the problem. The eleven hours after each one arrived were.
We run the whole chain for a small number of dental practices — the ads, the landing page, the response inside sixty seconds, the booking, and the cadence that protects it. Dental only, one practice per area, 10 on the roster.
Twenty minutes, your numbers rather than ours, and a plain answer about whether we can help.
Sikandar Habib, Founder, DENTALFLow Media
