It is nine thirty on a Friday night. Someone with a cracked molar searches "dentist near me", lands on your Google profile, and messages you on WhatsApp. You closed hours ago.
Monday at nine, someone on your team opens the conversation and replies. But that person is gone: they messaged two other practices the same Friday and booked with whoever answered first.
This is a model for working out what that costs you per year. It is not a success story. There is no client behind it, no observed results, and I am not promising you these numbers.
I publish it because dental marketing is full of "10 new patients guaranteed" without a single arithmetic operation behind it. I would rather show the method, mark what is hard data and what I made up, and let you judge.
First: where each number comes from
I start here because almost nobody publishes it, and because the difference between these three categories is the difference between an analysis and a brochure.
| Figure | What it is | Provenance |
|---|---|---|
| 90.6% of smartphone users in Mexico use instant messaging | Primary data | ENDUTIH 2025 — official survey by INEGI and IFT |
| Replying within an hour, rather than the next day, radically changes the odds of converting a contact | Published study | Harvard Business Review, 2011 — Oldroyd, McElheran and Elkington. Measured on B2B web-form leads, not dental patients |
| Treatment prices | Published price lists | Mexican dental clinics. An interested party, but it is what they actually charge |
| 20–40% of inquiries arrive outside office hours | Vendor data | Companies selling the exact fix for that problem. Tracing the 40% back, the source turned out to be the company itself, with no study behind it |
| 2–5 new-patient inquiries per day at a single-dentist practice | Vendor data | Dental software and consultancies. No dental association publishes an official figure |
| 40 messages a month · 2 lost · 1 closing a treatment | My assumption | None. These are numbers I chose, which is why this article asks you to change them |
If one thing here is useful, let it be this table. Next time someone shows you marketing projections, ask for exactly this. The answer will tell you more than the whole proposal.
Why I measure WhatsApp and not calls
Almost all published research on lost inquiries at practices measures phone calls, and it comes from the United States. In Mexico the channel is different.
According to INEGI's ENDUTIH 2025, instant messaging is the most widespread internet use in the country: 90.6% of the smartphone-using population. That is the most solid figure in this article, from a source nobody can dispute.
Now, carrying the missed-call problem over to the unanswered message is my inference. Reasonable, but an inference: I found no study measuring unanswered WhatsApp messages at Mexican dental practices. If one exists, send it to me and I will update this.
The model
A hypothetical practice in a metro area, two chairs, general mix with an orthodontics focus.
Treatment ranges come from public price lists of Mexican clinics in 2026:
| Treatment | Range |
|---|---|
| Metal bracket orthodontics | $18,000 – $35,000 MXN |
| Aesthetic or self-ligating orthodontics | $30,000 – $60,000 MXN |
| Invisalign-type aligners | $60,000 – $120,000 MXN |
| Single implant | from $32,000 MXN |
| All-on-4 restoration, per arch | $80,000 – $200,000 MXN |
For the calculation I take the floor of the cheapest treatment: one metal-bracket orthodontic case at $18,000. If your prices are higher — and they probably are — everything below improves.
The leak
Forty new messages a month. Dental consultancy benchmarks talk about two to five inquiries a day at a single-dentist practice. Forty a month sits at the low end of that range.
Thirty percent arrive outside office hours. This is where the most care is needed, and where I got it wrong in the first version of this article. The public figures range from 20% to 40%, and all of them come from companies selling software to answer those inquiries. A bigger number suits them. So I take the floor of the range, not the ceiling. That is twelve messages a month waiting until the next business day.
Of those twelve, two are genuinely lost. They book elsewhere before you reply. The other ten wait or message again.
And of those two, one would have closed a treatment. Not both. One.
| Item | Value |
|---|---|
| New messages per month | 40 |
| Outside office hours (30%) | 12 |
| Genuinely lost | 2 |
| Would have closed treatment | 1 |
| Treatment value | $18,000 MXN |
| Monthly leak | $18,000 MXN |
| Annual leak | $216,000 MXN |
One treatment a month. That is the entire assumption, and the annual leak still clears two hundred thousand pesos.
Why the result does not move
Lower the assumptions and you land in the same place. Thirty messages a month instead of forty? Still one treatment. Twenty percent after hours instead of thirty? Still one treatment.
The bottleneck is not volume: it is that one case is enough to move the needle when the ticket runs into five figures. This model only falls apart if you believe you never, not once a month, lose someone to a slow reply.
Break-even
A full stack — local SEO and GEO so they find you, plus an assistant that answers and books — costs around $18,980 MXN per month with $7,990 initial setup.
The first year is $7,990 + ($18,980 × 12) = $235,750 MXN.
At $18,000 per orthodontic case, that is 13.1 cases a year. A little over one a month.
Which is exactly what we estimate is being lost by not replying. Put differently: plugging the leak already pays for the whole model, and everything local SEO, GEO and reviews contribute sits above that line.
If your mix includes implants at $32,000 or restorations at $80,000, break-even drops to less than half.
When this model does not apply
No honest model gets published without its failure conditions. This one is no use to you if:
- You have no chair capacity. If you are already full, more new patients do not add: they create waiting lists and cancellations. Fix scheduling first.
- You already reply fast after hours. If someone is on WhatsApp in the evenings and at weekends, this particular leak is already plugged. The problem is elsewhere.
- Your in-chair close rate is low. I get the patient into the chair. Whether they accept the treatment plan happens in consultation, not in marketing.
- You get very few messages. At five a month instead of forty, the leak pays for nothing and what you need first is visibility, not automation.
- Your average ticket is low. A practice doing only cleanings and fillings has different arithmetic and probably a different plan.
If any of these is your case, I say so in the audit before you sign anything. A client who does not work out costs me more than one I never close.
Run it with your numbers
- Count last month's new messages. Do not estimate — count them.
- Check how many arrived after closing time or at weekends. That real percentage is worth more than any benchmark in this article, mine included.
- Be pessimistic about losses. Assume a handful slip away and only one would have closed.
- Multiply by your real ticket, not the market floor.
- Compare against the annual cost of the stack.
If the math does not work under conservative assumptions, do not force it. And if it works with room to spare, you know where to start.
This practice does not exist and represents no navamkt client or particular business. The volume and conversion assumptions are mine and are marked as such in the provenance table. If you spot an error in the math or a source, write to me: I will correct it and say so.