This clinic was handling 70 leads per month at Month 1. By Month 6, they were handling 85 leads per month, a 21% increase in volume. Revenue per lead went from €47 to €81. That €34 improvement, applied across 85 monthly leads, is €2,890 in additional monthly revenue from the same coordinator team, the same marketing spend, and the same clinical capacity. That is the EKSENAI deployment thesis made concrete: not more leads, better conversion on the leads already coming in.
Last Updated: 20260728T0
10 min read
A mid-size Istanbul dental clinic (veneers and implants focus, 60–90 leads/month, 4 coordinators) deployed EKSENAI in three phases over six months. Month 1 was baseline documentation. Month 2 activated the intake automation layer. Month 3 added the coordination layer. Months 4–6 added the management layer and full optimization. The data shows a 2.3x improvement in lead-to-consult rate, coordinator capacity freed from 80% inbound FAQ handling to 30%, and revenue per lead increasing by 71% by Month 6.
I’ve built intake systems for clinics across hair transplant, dental, and cosmetic surgery. This case study covers the full six-month deployment arc for a mid-size Istanbul dental clinic, veneers and implants primary focus, 60–90 incoming leads per month, four coordinators, no prior CRM infrastructure. All identifying information has been removed at the clinic’s request, but the metrics are real. I’m sharing them because the pattern is consistent enough across deployments that it is more useful as documentation than as a sales pitch.
| Metric | Month 1 (Baseline) | Month 3 (Post-Intake Layer) | Month 6 (Full System) | Change |
|---|---|---|---|---|
| Monthly leads (inbound) | 70 | 74 | 85 | +21% |
| Average lead response time | 4.2 hours | 12 minutes | 8 minutes | -97% |
| Lead-to-consult rate | 18% | 31% | 41% | +128% |
| TFCR (Treatment-to-First-Contact Rate) | 6.1% | 9.4% | 13.8% | +126% |
| Coordinator hours on FAQ/inbound handling | 80% of capacity | 52% of capacity | 30% of capacity | -63% |
| Revenue per lead (€) | €47 | €64 | €81 | +72% |
| Average monthly procedure revenue (est.) | €67,200 | €91,200 | €131,800 | +96% |
The TFCR: Treatment-to-First-Contact Rate, is the metric that most clinic operators have not heard of and that best predicts the financial return on intake improvements. It measures what percentage of first-contact inquiries convert to a booked and attended procedure. At 6.1%, this clinic was converting 6 patients for every 100 leads that made first contact. By Month 6, they were converting 14. The same marketing spend, the same lead volume class, and nearly 2.3x the procedures booked.
What Did the Baseline Audit Actually Find?
The Month 1 baseline audit was a three-week process: reviewing 90 days of lead data (source channel, inquiry date and time, first response timestamp, outcome), shadowing coordinator workflows, auditing the consent documentation, reviewing the pricing communication process, and mapping the WhatsApp inbox structure.
What the audit found was a pattern I’ve seen repeatedly in Istanbul clinics of this size. The four coordinators were each managing personal WhatsApp inboxes with 30–50 active lead conversations simultaneously. There was no CRM. There was no shared pipeline view. When a coordinator was out sick, their leads were either missed entirely or another coordinator tried to context-switch into an unfamiliar conversation thread. The lead response time of 4.2 hours was a business-hours average, inquiries arriving after 7pm or on weekends were typically picked up the following morning.
Of the 70 monthly leads, approximately 35% were being lost at the first-response stage: they never received a timely, useful first response and had moved on before a coordinator engaged. Of the 65% that received a response, approximately 28% converted to a consultation call. Of those, approximately 33% booked a procedure. The compounding effect of three conversion failures, first response rate, inquiry-to-consult rate, consult-to-booking rate, produced the 6.1% TFCR.
The coordinator capacity audit found that 80% of coordinator time was consumed by what I call FAQ volume: questions that are standard across every international dental patient — “what is included in the package?”, “what hotel do you use?”, “how many nights do I need to stay?”, “do you accept credit cards?”. These questions do not require a coordinator’s clinical or commercial judgment. They require a knowledge base.
How Did the Three Deployment Phases Actually Work?
Phase 1: Intake Layer (Month 2)
The intake layer deployment had three components. First, the knowledge base: 25 hours of coordinator and clinic manager time over two weeks to build a five-layer Supabase knowledge base covering the clinic’s procedure library (veneers, implants, composite bonding, all-on-4), surgeon profiles, logistics FAQ, pricing tiers, and objection library. Second, the Evolution API WhatsApp integration: connecting the clinic’s main WhatsApp number to the n8n intake workflow, so all incoming inquiries entered a structured pipeline rather than landing in a personal inbox. Third, the automated first-response layer: a 90-second acknowledgment message sent to every new inquiry, a RAG-backed AI response system that could answer 70–80% of FAQ volume from the knowledge base, and a human escalation trigger for inquiries involving pricing negotiation, clinical assessment requests, or explicit high-intent signals.
The impact on lead response time was immediate: from 4.2 hours average to 12 minutes average within the first two weeks of Month 2. The AI handled 68% of initial FAQ volume without coordinator involvement. Coordinator time on FAQ handling dropped from 80% to 52% of capacity. Lead-to-consult rate improved from 18% to 31%, primarily because leads were receiving useful, specific information within minutes of inquiry instead of waiting hours for a generic coordinator response.
Phase 2: Coordination Layer (Month 3)
With the intake layer stabilized, Month 3 focused on the coordination layer: implementing Chatwoot as the CRM, migrating all active leads from personal WhatsApp inboxes to structured Chatwoot conversations, implementing a lead qualification scoring system in Supabase (source country, procedure interest, timeline, previous clinic contacts), and building the post-deposit upsell workflow in n8n.
The Chatwoot migration was the most operationally disruptive phase. Coordinators who had built their entire working process around personal WhatsApp management resisted the transition. The resistance was not malicious, it was the natural friction of changing a habitual workflow. The transition required two weeks of parallel operation (both personal WhatsApp and Chatwoot active simultaneously), followed by a hard cutover enforced by clinic management.
The coordination layer improvements produced: full pipeline visibility for clinic management (for the first time, the clinic director could see every active lead, their status, and the last touchpoint date), automated follow-up sequences for leads that had gone cold (6 and 14 days without coordinator response triggered automated re-engagement messages), and post-deposit upsell workflow that generated an average €140 in additional revenue per converted patient during Month 3.
Phase 3: Management Layer and Full Optimization (Months 4–6)
The management layer deployment covered: revenue analytics dashboard in Supabase (revenue per lead by source channel, coordinator conversion rates, procedure category mix), automated review solicitation at 14 days post-discharge, a WhatsApp referral program triggered at 30-day post-op check-in, and knowledge base expansion based on the FAQ data accumulated in Months 2–3.
The revenue analytics dashboard surfaced a finding that the clinic director had not previously quantified: the clinic’s Instagram-sourced leads had a 2.1x higher revenue per lead than their Google Ads-sourced leads, but the clinic’s marketing spend was weighted 70% toward Google Ads. Reallocating 30% of marketing spend from Google Ads to Instagram content in Month 5 increased average monthly revenue per lead without increasing total marketing spend.
By Month 6, the TFCR had reached 13.8%, coordinator capacity on FAQ handling had dropped to 30% (freeing the coordinator team to focus on consultation quality and post-deposit relationship management), and revenue per lead had reached €81.
What Is the Underlying Principle Most Turkish Clinic Operators Miss?
Revenue leakage in a dental clinic operating at this size is not a mystery. It is measurable. The 6.1% TFCR in Month 1 was not bad luck, it was the predictable outcome of a 4.2-hour average response time, a coordinator team spending 80% of their capacity on FAQ volume, no pipeline visibility, and no systematic follow-up for leads that went cold. Every one of those failure points had a specific, quantifiable revenue cost.
The operational insight from this deployment is that intake system investment at this scale, approximately €3,200 in total deployment cost across infrastructure, knowledge base development, and Chatwoot licensing, generated a revenue increase of approximately €64,600 in the six-month period compared to the extrapolated baseline trajectory. The ROI calculation is straightforward. The investment barrier is not financial. It is the belief, common among Istanbul clinic operators, that their conversion problem is a marketing problem, that they need more leads rather than better systems for the leads they already have.
Frequently Asked Questions
How long did the coordinators take to adapt to working in Chatwoot instead of personal WhatsApp?
Full adaptation, meaning the coordinators defaulted to Chatwoot rather than personal WhatsApp for all patient communication, took approximately six weeks from the hard cutover date. The first two weeks after cutover involved regular corrections as coordinators reverted to personal WhatsApp threads for new inquiries out of habit. Weeks three and four saw consistent compliance for new leads but continued personal-phone management of older active conversations. By week six, the transition was complete for all active leads. The primary behavioral driver that accelerated adoption was the management-layer analytics dashboard: coordinators became aware that their individual conversion rates were visible to the clinic director, which created a performance accountability environment that the personal-WhatsApp model had made impossible.
What was the coordinator team’s reaction to the AI handling initial patient inquiries?
Mixed initially, settled positively by Month 3. The initial concern was that patients would perceive the automated response as impersonal and that it would damage the clinic’s reputation for service quality. The data from Month 2 resolved this concern: the AI response quality from the structured knowledge base was consistently rated as helpful and informative in patient feedback surveys, and the 90-second response time generated positive comments in several early post-deployment reviews. The coordinators’ concern shifted from “patients will not like the AI” to “when does my role start in the conversation”, which is a productive question with a clear operational answer (human coordinator engagement triggers on high-intent signals, pricing negotiation requests, and clinical assessment questions).
What was the single highest-impact change in the deployment?
By data, the highest-impact single change was the 90-second automated first-response with RAG-backed FAQ handling. The lead-to-consult rate improvement from 18% to 31% between Month 1 and Month 3 was primarily attributable to this change, patients who previously received no response within the decision window were now receiving specific, useful information within minutes. The ROI per hour invested in the knowledge base build (approximately 25 hours of staff time) was higher than any other component of the deployment.
Did the upsell workflow produce meaningful revenue in the first deployment months?
Yes, though it was not the primary revenue driver in the first 90 days. The post-deposit upsell sequence generated an average €140 per converted patient in Month 3, rising to €195 by Month 6 as the upsell offerings were refined based on acceptance data. For a clinic converting 13–15 patients per month by Month 6, this represented €2,535–€2,925 in monthly upsell revenue, material, but smaller than the revenue gain from the TFCR improvement. The upsell revenue is more visible as a percentage contribution in clinics with a higher average procedure value, where the upsell-to-primary ratio is more favorable.
Could a smaller clinic with 20–30 leads per month justify this deployment?
Yes, with a modified deployment scope. For a clinic at 20–30 leads per month, the full Chatwoot + n8n + Supabase + Evolution API stack is appropriate infrastructure but the knowledge base development and automation sophistication can be simplified. The core high-ROI components — 90-second automated first response, basic knowledge base for FAQ handling, and a simple follow-up sequence for cold leads, can be deployed in approximately 40 hours of build time versus the 80–100 hours required for the full deployment. The TFCR improvement at this lead volume is typically smaller in absolute revenue terms but proportionally similar, and the coordinator capacity freed from FAQ handling is equally valuable in a smaller team.
[Reviewed by Dr. Hasan Şahin, Medical Director at MedTurkAI]
*Running a clinic and not sure where your pipeline is leaking?*