Implantology in Odesa: lead-to-payment conversion from 40% to 68%

The clinic is narrowly specialized — dental implantation. It runs 2 surgeons, both deep in the surgery flow, which means their time is expensive and spending it on phone calls to clients is a direct loss of money. There's no separate sales team: leads went through an administrator who was simultaneously running the lounge.

The key difference from regular dentistry is that this isn't a "UAH 1,200 cleaning booking", but a deal of tens of thousands of hryvnia with a long decision cycle. People don't pay for implantation on the first evening. First — a free consultation and CT scan, then an individual treatment plan, then deliberation ("I'll think about it", "I'll consult my husband", "I'll compare with another clinic"), and only then — a deposit and the surgery itself. It's a classic sales funnel for an expensive service, where it's easy to lose the client between stages.

Before MTDK ai there was no systematic nurturing of the lead between stages. The administrator booked the consultation, the surgeon did the exam, then the plan was printed or dictated verbally — and the client went off "to think". Who would call them back and when, whether anyone would remind them about the deposit or answer questions about implant materials — depended on whether someone got around to it. Often no one did.

The baseline we recorded before launch, to have something to compare against: lead-to-payment conversion around 40%, an average deal cycle (from first contact to deposit) of 2–4 weeks, on average ≈3 weeks, and a share of leads that reached a treatment plan at all of roughly 55%.

Implantology is the most profitable niche for automation precisely because what's at stake here isn't a missed UAH 1,000 booking, but a deal that can pay back a month of AI from a single saved person. So we built not a "dental chatbot in the classic sense", but an AI clinic administrator sharpened for one task — converting the lead into a payment.

All three pains had a single root: no one systematically "nurtured" the expensive deal between stages, and an unsupported lead cools off — especially when it's about tens of thousands of hryvnia that people are afraid to part with.

Gap one — the drop between plan and deposit. The client got the treatment plan, said "thanks, I'll think about it" — and vanished. No one reminded them about the deposit, brought them back to the conversation, or addressed the doubts that ripened at home. The owner estimated that most "warm" leads who genuinely wanted implants were lost at this stage, simply for lack of a nudge at the right moment.

Gap two — questions about materials and the warranty went unanswered in time. Before paying tens of thousands, a person wants to understand: which implant systems, the price difference between them, the warranty, what happens if the implant fails to integrate. The administrator couldn't always answer confidently, and pulling a surgeon away for every question was awkward. Silence at this moment = a lost deal.

Gap three — surgeons spent their time on calls and follow-up. When the owner did ask the surgeons to personally "close" a warm client, they were torn away from surgeries for phone conversations. It's the most expensive possible way to nurture a lead: an operating doctor playing sales manager.

The heart of the solution is an AI funnel manager in Telegram that guides every lead through clearly defined funnel stages: consultation → plan → deposit → implantation. It doesn't "book an appointment" in the usual sense — it nurtures an expensive deal from the first "I'm interested in implants" to the moment the client pays a deposit and is ready for surgery.

The knowledge base: the lineup of implant systems with an honest comparison by price and lifespan, the warranty terms and what happens if integration fails, common objections ("too expensive", "I'll think about it", "I'll consult") with rehearsed responses, and clear boundaries — the AI doesn't make diagnoses or promise medical outcomes, that's always the surgeon's domain.

A separate lever is the treatment plan in PDF. After the consultation the AI builds a personal treatment plan as a tidy PDF: the stages, the chosen implant system, the estimated cost, the timeline and the warranty terms. The client receives it in Telegram right away — not "in a couple of days, when the admin finds time", but the same evening, while the impression of the consultation is still fresh. The PDF is easy to forward to the spouse the person consults with — and that, too, is part of the funnel.

The key scenario is gentle deposit reminders. The AI doesn't push in a "so, when are you going to pay" style. It returns to the client with care: whether any questions remain after reviewing the plan, reminds them that the surgeon's slot is reserved with a deposit, and offers to pay it online via LiqPay right in the chat. Exactly this careful but persistent escort between "got the plan" and "paid the deposit" closed the funnel's main gap.

Handing off to a surgeon — only when the client is ready. The AI doesn't pull doctors in for every question. It holds the client through the entire deliberation stage on its own and hands them to a surgeon with full context only when they're actually ready for surgery: a deposit is paid, or the person explicitly asks to schedule a date. Technically it all runs on n8n — it ties Telegram, PDF generation and LiqPay into a single automatic scenario.

Better shown live than described. Here's a typical inquiry the day after a consultation — the moment where the client used to stall for a week, and now the AI nurtures them toward a deposit, answering real questions about materials and the warranty:

No pressure and no "when are you paying". The client gets concrete answers about implant systems and the warranty, sees their plan in PDF, understands that the slot is reserved with a deposit — and makes the decision themselves, but no longer alone with their doubts. And the surgeon steps in only at the final step, once everything is decided.

Days 1–3. We break down the funnel stage by stage with the owner: where exactly clients are lost, how the plan is currently created, which objections come up most often. We collect the implant lineup, warranty terms, price list. This is the foundation of the prompt.

Days 4–6. Setting up the AI for selling, not booking: objection handling, a tone without pressure, boundaries (where the manager ends and the surgeon begins). In parallel — the treatment-plan PDF template and its automatic generation.

Days 7–9. Connecting LiqPay for an online deposit right in the chat and assembling the whole scenario in n8n. Testing on 40 internal dialogs — we log every client "stall" and tune the reminder logic.

Days 10–11. Launch on real leads in "AI nurtures, owner sees every dialog" mode. For the first days we watch the deposit moment closely — it's the most sensitive point of the funnel.

Day 12+. Full autonomy. The AI nurtures the lead from consultation to deposit on its own, and the surgeons receive a client already ready for surgery with full context — and not a single call on their side.

Two and a half months after the full launch we tallied up the numbers. These aren't marketing roundings — they're data from the funnel CRM, the Telegram bot statistics and LiqPay payments:

Lead-to-payment conversion: 40% → 68%. This is the case's headline number. For every 100 leads the clinic now brings 68 to a deposit instead of 40 — and these aren't new leads or extra ad budget, but the very same inquiries that used to cool off after the plan.

The deal cycle shrank from ≈3 weeks (2–4) to 8 days. The client stopped "stalling": the plan arrives in PDF that same evening, questions about materials and the warranty get an instant answer, and gentle reminders keep the decision from dragging out for a month. The faster the cycle, the lower the chance the person changes their mind or leaves for another clinic.

The share of leads reaching a treatment plan grew from 55% to 82%. Previously some people dropped off even before getting a plan — not waiting for a reply or getting lost after the consultation. Now the AI brings almost everyone who came for a consultation to a plan.

Surgeons are fully freed from calls and follow-up. Before launch they occasionally "closed" warm clients personally, torn away from surgeries. Now they step in only at the final step — for a client who has already paid a deposit. The operating doctor's time returned to where it earns money — the chair.

We separately calculated the economics of one saved lead: with an average deal of tens of thousands of hryvnia, even one extra closed deal a month covers the cost of the AI with room to spare. And a 28-point conversion lift isn't one saved deal — it's a systematic flow.

Let's be precise: the AI doesn't operate and doesn't "sell instead of the clinic" — it closed exactly the stage that used to sag, nurturing the lead between consultation and deposit. Here's how the key funnel points changed before and after launch:

For the first days the AI reminded about the deposit too insistently — clients felt the pressure typical of cheap "hard closes". For an expensive medical service that scares people off. We rewrote the logic: care and objection handling first, deposit reminders only after the client confirmed the plan works for them.

At first the AI tried to explain clinical nuances itself ("in your case a two-stage implant is better"). That's unacceptable — any decision about the treatment method is the surgeon's. We fixed the prompt with a hard rule: anything touching a medical decision → "the surgeon will determine this at the consultation, I can book you in".

The first version of the PDF plan was overloaded with terms — it scared the client rather than reassuring them. We simplified it: clear language, distinct stages, transparent cost and warranty without fine print. A plan should sell through confidence, not confuse.

Initially the AI offered full prepayment upfront — and that repelled people on an expensive deal. We switched to a deposit that reserves the surgeon's slot: psychologically it's easier to pay a part, and from there the client is already "in the process". The takeaway is simple: in selling an expensive service, pace and tact matter more than speed of closing.

Now the main question of any owner — how much it costs and when it pays off. For this clinic we used the "Business" plan — €150/mo (≈UAH 7,700) plus a one-time setup of €600 (≈UAH 30,800): the funnel scenario with the PDF plan and payments is more complex than the basic one, but there are fewer integrations than at chain clinics, so a mid plan rather than the top one.

The economics here are counted differently from regular dentistry, because the tally isn't in the number of bookings but in closed deals. A conversion lift from 40% to 68% means that from the same lead flow the clinic closes noticeably more expensive deals. With the average implantation ticket, even a few extra deals a month is hundreds of thousands of hryvnia in turnover that used to simply leak out of the funnel after the plan stage.

Payback of the full package (subscription + setup in the first month) — about 9 days: effectively one or two extra closed deals already cover all the AI costs for the year. From the second month on, the monthly fee pays for itself from a single saved consultation. For comparison: hiring a dedicated sales manager to systematically nurture every lead in the evenings and on weekends would cost several times more — and they still wouldn't generate a PDF in seconds or remind hundreds of clients at once.

The clinic owner's bottom line: "we thought the problem was the number of leads, but it turned out we just weren't bringing them to payment". When clients happily come for a consultation, get a plan — and disappear, the question isn't demand but the step between "interested" and "paid". Closing that gap took 12 days.