A clinic conversation that’s connected to the clinic itself.
Menaela communicates with patients and leads through connected channels such as WhatsApp, using live clinic information instead of a static FAQ bot. Conversations are bound to the right patient or lead and to the real schedule and catalog.
A patient message is rarely just a message. “Can I come Thursday?” needs identity, provider rules, duration, availability and a booking action. “How much is it?” may depend on service, provider and payer. Menaela keeps that context inside the clinic system, so the conversation can finish the job instead of producing another task for reception.
Not a demo prompt. These are product capabilities connected to the clinic’s actual data and workflows.
WhatsApp conversations
Receive and send clinic WhatsApp messages, keep a unified dialog history and connect each conversation to the relevant patient or lead.
Identity-aware conversations
Unknown inbound contacts can enter the lead flow; known patients stay bound to their patient identity, with server-side conversation context carrying the person and clinic boundary forward.
Context retrieval from clinic data
Answer operational questions from the configured clinic catalog, provider data, pricing, schedule and the permitted parts of the patient record instead of relying on a static FAQ prompt.
Appointment management in chat
Patients ask for times, pick a slot, book, reschedule, cancel or confirm — in the channel they already use.
Human escalation & takeover
Requests needing clinical judgement, policy exceptions, uncertain identity or unsupported actions are routed to staff; staff can take over a conversation and hand it back to Menaela with the conversation context preserved.
Per-channel autonomy
Enable autonomous handling separately for WhatsApp, email leads, Meta lead forms and the Patient Chat Test instead of exposing every connected channel at once.
Known-patients-only WhatsApp
Restrict WhatsApp autonomy to recognized existing patients. Unknown senders stay outside the autonomous AI flow and are not auto-created as leads, while manual messaging still works.
What the capability looks like when the request is slightly messier than a feature list.
Patients change the question halfway through.
The conversation keeps the service, provider and offered slots in context while still respecting clinic-specific authority boundaries.
I need a cardiologist sometime this week.
I can offer Dr Levi Wednesday at 17:30 or Dr Cohen Thursday at 18:00.
Thursday is better. Anything earlier?
Dr Cohen also has 16:30 on Thursday.
How much is it with Maccabi?
The operational path behind a normal request.
A message arrives from a known patient or a new lead.
Menaela binds the conversation to the correct person and uses the clinic catalog, schedule and permitted patient-facing operations.
It answers routine operational questions or offers real appointment options.
When judgement or a non-routine decision is required, the conversation is handed to staff with its context intact.
How the capability is bounded in the product, not just what appears in the interface.
Conversation identity is server-owned
The model receives a conversation that is already bound to the clinic and, where known, to the patient or lead. Identity is not selected by free-text prompt instructions.
Context is retrieved on demand
The patient-facing assistant requests only the operational context needed for the current message — for example service data, pricing, bookable times or approved recorded-care information.
Failure is a valid outcome
If the clinic has not configured an answer, identity is uncertain or a tool fails, the assistant should escalate or state the limitation rather than manufacture a plausible answer.
Human takeover preserves the thread
Staff can continue the same conversation with the message history and current operational context visible instead of starting a second support channel.
Natural-language requests Menaela can act on.
- Connected to the real schedule, patient identity and clinic catalog.
- No separate chatbot knowledge base needed for basic operational facts.
- Patient-facing tools are deliberately narrower than staff tools.
- Patient identity is server-bound; the conversation cannot simply name another internal patient ID and switch context.
- A connected personal or mixed-use WhatsApp can run in known-patients-only mode, keeping unknown private contacts outside autonomous AI handling.
- The assistant is expected to say when information is unavailable or an action cannot be completed, rather than inventing clinic facts.
- Clinics where WhatsApp is already the de-facto front desk
- Teams losing leads outside office hours
- Practices that want booking and communication to share one source of truth
Patient-facing autonomy is intentionally narrower than staff access. Clinical judgement, unrestricted chart access and sensitive internal operations do not belong in the external toolset.
Deeper explanations of the workflows and edge cases behind this capability.
Ready to hand over your practice management?
One fraction of the cost. Every hour of the day. The same calm voice at midnight as in the morning.
Start with Menaela