What an AI receptionist actually does in a clinic

The term "AI receptionist" covers two distinct things: a voice assistant that handles phone calls and a chatbot that manages conversations via WhatsApp, the clinic website, or both channels simultaneously. The most complete solutions combine them.

In practice, a well-configured system can do the following without human intervention:

  • 24/7 availability — answers calls and messages outside clinic hours, including evenings, weekends and public holidays.
  • Appointment management — books, modifies and cancels appointments by checking real-time availability against the clinic calendar.
  • FAQs — provides information on indicative prices, accepted insurance, location, opening hours, pre-treatment preparation and other standard queries.
  • Automated reminders — notifies the patient 24 or 48 hours before their appointment and lets them confirm or reschedule directly from the message.
  • Multilingual service — detects the patient's language and responds accordingly. Where Spanish, British, German, Russian and Nordic patients coexist, this has direct practical value.
  • Escalation to a person — when it detects a query requiring clinical judgement, a genuine emergency or a patient who asks to speak with someone, it transfers the conversation to the human team with full context already visible.

What it cannot do: give diagnoses, interpret symptoms, make clinical decisions or replace a healthcare professional in any circumstance. Any provider who claims otherwise should be treated with scepticism.

Inbound contact flow

01Call / WhatsAppPatient contacts outside opening hours or during the day.
→
02AI ReceptionistDetects intent and responds instantly 24/7.
→
03Books appointmentChecks real availability and confirms the slot.
→
04Escalates to staffIf urgent or sensitive, transfers with full context.

Simplified flow — the AI resolves bookings, FAQs and escalations in parallel

When an AI receptionist does make sense

There are three clear signals that a clinic is ready to benefit from this technology:

1. Unmanageable contact volume

If reception spends more than two hours a day answering routine calls and messages — appointment confirmations, price enquiries, opening hours — that time has a real cost. A dental clinic with 80 or 100 patients per week, or a physiotherapy clinic with back-to-back sessions, is typically at this threshold.

2. Missed calls outside opening hours

Most patients call when they can, not when the clinic is open. If the last consultation slot ends at 8pm but the patient only has a free moment at 9:30pm, that call is lost. In an aesthetic or dental clinic, one missed appointment can represent a sizeable loss depending on the treatment — from a few dozen to several hundred euros in the higher-value cases. A system that captures those out-of-hours requests pays for itself quickly when volume is sufficient.

3. High no-show rate

Unannounced absences are the most costly problem for many clinics. In the private healthcare sector, no-show rates without active reminders can fall in the range of 10% to 20% depending on the type of centre and specialty. With well-configured automated reminders, that number drops significantly in most cases — though the specific result depends on patient profile and how the reminder flow is designed. There is no universal figure.

Tourist-area context: many clinics routinely see patients in Spanish, English and German, and during high season also in Russian, French and Scandinavian languages. An AI receptionist that responds in the patient's language removes the communication barrier that often causes a tourist to give up on booking an appointment.

AI vs human team

AI handles

  • ✓Books, modifies and cancels appointments
  • ✓Prices, hours and standard info
  • ✓Reminders and confirmations
  • ✓Multilingual service 24/7

Passed to staff

  • →Clinical emergencies
  • →Complex medical queries
  • →Patient asks to speak to someone
  • →Out-of-scope situations

Task split in a well-configured clinic

When it doesn't make sense (or not yet)

Honesty here matters, because the technology has setup and configuration costs that aren't always justified:

  • Low call volume. If reception receives 10 to 15 calls a day and most are manageable in the first 30 minutes after opening, the impact of an AI assistant is marginal and the return will take time to materialise.
  • Where human contact is part of the value proposition. Some high-end clinics build their reputation on personal care from the very first interaction. In those cases, automating reception may work against the experience the patient expects.
  • No practice management software. An AI receptionist needs to connect to a calendar or management system (Doctoralia, Gesden, Clinicoud, etc.) to operate. If the clinic manages its schedule on paper or in an unshared spreadsheet, implementation requires first digitising internal management.
  • High-complexity specialties in the first contact. In psychology, oncology or internal medicine, the first call is often a sensitive conversation requiring active listening. A poorly designed automated system can damage the clinic's reputation at that first touchpoint.

Does it make sense for your clinic?

Act now

strong fit

+30 contacts/day · No-show rate ≥10% without reminders · Connected scheduling software (Doctoralia, Google Calendar) · Missed calls outside opening hours.

Wait

not yet

Fewer than 15 calls/day · Paper or spreadsheet schedule · No-shows already below 5% · First contact is highly sensitive (psychology, oncology).

Assess your clinic before speaking with any provider

How much does an AI WhatsApp receptionist cost, in a clinic or any other small business?

Ours has a fixed, published price: 990 € to set up and 79 €/month for maintenance, no contract term, VAT not included, and third-party APIs (WhatsApp Business API, the AI model) billed by their providers. It is the same rate for a clinic, a workshop, a shop or an accountancy: the listing is in the catalogue and it can be seen working on video before you buy.

What exactly does it do for that price?

  • Runs through Meta's official API: they message your business WhatsApp and your business answers, not a generic bot.
  • Checks your catalogue before replying and answers with product, price and link.
  • Every conversation lands in the CRM.

What is not included?

  • The WhatsApp Business API: Meta charges per conversation, directly to you. We help you set it up, and we explain how it differs from the regular app in WhatsApp Business API vs. regular WhatsApp.
  • The AI model: usage is billed by its provider.
  • The CRM: if you have none, we suggest one; if you do, we connect to it.
  • Major flow changes after delivery: quoted separately.

Why a fixed price and not "from"?

Because a "from" that balloons at the end does not help anyone decide. The 10 fixed-price automations in the catalogue all have that price published; only bespoke projects (a website, an online shop) carry "from", and we explain why.

How do you start?

  1. Free diagnosis of your online presence with the Zerolagia scanner (0 €).
  2. A call to see what questions you get and what you want the assistant to answer.
  3. Setup, testing with your real messages, and handover with access for your team.

What the price covers and what is billed separately

Covered by 990 € + 79 €/month

  • ✓Setup with your data and your answers
  • ✓Catalogue lookup before replying
  • ✓Every conversation saved in the CRM
  • ✓Monthly maintenance, no contract term

Billed separately

  • →WhatsApp Business API, charged by Meta
  • →AI model usage
  • →The CRM, if one has to be bought
  • →Major flow changes after delivery

Fixed catalogue price — VAT not included, no contract term

GDPR compliance: what cannot be ignored

Health data is a special category under the General Data Protection Regulation. That means not every tool qualifies: there are specific requirements that must be met before deploying an AI assistant in a clinic.

Key points to verify with any provider:

  • Where data is processed. Data must be processed within the European Union or, at minimum, under a valid international transfer framework. Tools that process data on US servers without additional safeguards are problematic for health data.
  • Data Processing Agreement. The provider must sign a DPA that clearly establishes how patient data is handled.
  • Use of data for training. Some providers use conversations to improve their models. With health data, this requires explicit patient consent, which in practice is very difficult to obtain. The alternative is to contract a service that expressly guarantees it does not use data for this purpose.
  • Records of processing activities. The clinic must update its processing records to include the new data processing activity the AI assistant introduces.

This is not an insurmountable obstacle, but it is a step many clinics skip — which can lead to sanctions from the Spanish data protection authority (AEPD). It is worth verifying before signing any contract.

GDPR checklist before signing

  • ✓Data processed within the EU
  • ✓DPA signed with the provider
  • ✓Provider does not train on your data without consent
  • ✓Records of processing activities updated
  • ✓Privacy notice updated on the clinic website

Not optional — health data = special category (GDPR art. 9). Non-compliance → AEPD fine

How to assess whether now is the right time for your clinic

Before speaking with any provider, it is worth answering these four questions:

  1. How many calls or messages does the clinic receive on a normal day? If fewer than 15, the impact will be limited. If more than 30, there is likely time and money to recover.
  2. How many appointments are lost because people call outside opening hours? If you don't have this figure, run the test for a week: note every missed call that doesn't get returned. The number tends to be surprising.
  3. What is the current no-show rate? If it is low (below 5%) and manual reminders are already working well, the margin for improvement from an automated system is smaller.
  4. Do you have internet-connected practice management software? Without this, integration is significantly more complex and expensive.

Compatible integrations

Doctoralia, Gesden and Clinic Cloud connect via export or middleware — no need to migrate your current software

Is your clinic switchboard overwhelmed?

At Zerolagia we analyse your clinic's specific situation before recommending anything. If an AI assistant makes sense, we show you how and how much. If it doesn't make sense yet, we tell you that too.

Let's talk — no commitment

Frequently asked questions

Does the AI receptionist integrate with Doctoralia or Gesden?

It depends on the solution. Doctoralia has an API and allows external tools to connect for calendar management. Gesden, being desktop software, requires a middleware layer (webhook or custom integration). Both are technically feasible, but need to be assessed case by case before committing to any solution.

Is the AI receptionist GDPR-compliant?

It must be, but not all solutions on the market guarantee this by default. Health data is a special category under GDPR and requires reinforced measures. You must verify that the provider processes data within the EU, signs a Data Processing Agreement, and does not use the data to train its models without explicit consent.

What languages can the AI receptionist handle?

Modern systems built on language models can handle Spanish, English, German, French, Russian and other languages common among international customers, provided they are configured correctly. Patient language detection is automatic in most solutions.

What happens when a case is complex or a patient needs to speak with someone?

A well-configured system detects when it cannot resolve an enquiry — because it is urgent, requires clinical judgement, or the patient explicitly asks to speak with a person — and escalates to the reception team or relevant professional. The patient receives confirmation that someone will be in touch, and the team receives the full conversation context.

How long does it take to get the AI receptionist up and running?

For a small or medium-sized clinic, the typical timeline is two to four weeks from the decision to implement. It depends on the integration with the existing practice management software and the configuration of the conversation flows specific to the centre.

Do I need a new WhatsApp Business account?

You need the WhatsApp Business API, which can be your current number once migrated. We do it with you.

Can it sell or book appointments?

It can capture the request and put it in the CRM. Booking the visit is a separate automation in the catalogue, and there nothing is booked until you confirm it.

What if the AI doesn't know something?

It says so and passes the conversation to a person: when it doesn't know something, it tells you it doesn't know.

Is there a contract term?

No. You pay the setup and the monthly fee for as long as you use it.

You might also like

Automation for clinics and health centres Free website scanner WhatsApp receptionist with AI: listing and price All articles