An AI receptionist that books into your real calendar.
We build AI receptionists that answer website chat, text messages and phone calls, book and reschedule appointments in the scheduling system you already run, answer from your own approved information, and pass anything else to a person with a summary. Built, integrated and deployed for one flat fee, from $8,000.
What an AI receptionist should do, and what it should not.
An AI receptionist handles the front-desk work that repeats all day: opening hours, directions, prices, insurance and preparation questions, new bookings, reschedules and cancellations. It is only useful if three things hold. It reads live availability and writes the booking into the system your staff use, so nobody re-keys anything. It answers from information you approved, so it does not make up a policy or a price. And it knows when to stop, passing complaints, clinical questions and anything unusual to a person with the conversation summarised. Gigabit builds all three, with every conversation logged and reviewed.
Two published healthcare builds show the parts. A two-location dermatology clinic's patient assistant answers administrative and prep questions on a web widget, SMS and WhatsApp, with phone-menu deflection to self-serve and after-hours replies; in 8 weeks repetitive front-desk inquiries fell 62%, median time-to-answer on common questions went from 4:10 to 1:20, and 40% of after-hours interactions resolved without staff. A six-location dental group's self-serve booking syncs in real time with Dentrix, and two-way SMS and WhatsApp reminders let patients confirm or reschedule; within 60 days kept appointments rose 27% and no-shows fell 29%. Neither build was an AI voice agent holding a spoken phone conversation. We build voice receptionists too, and we will tell you that we have no published voice case yet.
What an AI receptionist build includes.
Each build ships with booking rules, an approved knowledge source, escalation paths, conversation logs and a dashboard of what it handled.
Website chat, SMS and WhatsApp
One receptionist across the site widget and messaging channels, so after-hours messages get an answer instead of waiting in a morning backlog.
Phone answering
A voice receptionist that answers calls in natural speech, or a phone menu that sends routine callers to self-serve, with a transfer to staff for anything else.
Booking in your real system
Live availability read from your scheduling, practice-management or booking system, bookings and reschedules written back, with rules for visit types, durations and providers.
Answers from your own information
Hours, services, prices, policies and preparation instructions from a source your team approves and updates, with no answers outside it.
Handoff to a person
Keyword and intent triggers route complaints, urgent and clinical questions to staff, with a summary so nobody asks the customer to start again.
Privacy and audit trail
Only the fields each task needs, encryption in transit and at rest, logged conversations and actions, and a BAA chain where health information is involved.
From the front desk's busiest calls to a receptionist that handles them.
Map the calls
List what customers ask and book, from call logs, inboxes and staff, and agree what the receptionist answers, books, refuses and hands over.
Connect the calendar
Wire it to your scheduling system with booking rules and conflict handling, and test it against real availability before anyone talks to it.
Launch on one channel
Start with website chat or texts, review every conversation for the first weeks, then add channels and phone answering.
Review and improve
Weekly transcript reviews, updates to the approved answers, and a dashboard of handled, booked and escalated conversations.
Published prices, mapped to the job.
Every price here is the same one on the pricing page. Work with no published price gets a fixed price in writing before it starts.
Figures in the notes are what past clients paid, from the published case studies; they show the range, not a rate card. Work with no natural end, such as ongoing development, is priced per engineer per month with no lock-in.
When to hire us for this, and when not to.
- Calls and messages go unanswered at busy times or after hours, and some of those are lost bookings.
- Most front-desk questions repeat, and the answers can be written down and approved.
- Bookings live in a scheduling or practice-management system that can be read and written to.
- Someone can own the approved answers and review conversations each week.
- Your booking tool already has a self-serve page and reminders you have not switched on. Use them first.
- Callers mostly need judgement a receptionist should not exercise alone, such as clinical triage or legal advice.
- You want a receptionist that answers anything, from any source. One that stays inside approved answers is the one worth running.
Two front-desk builds, with the numbers.
A clinic assistant on web chat, SMS and WhatsApp, and a dental group's self-serve booking and two-way reminders synced with Dentrix.
A RAG patient assistant with clinical guardrails cut repetitive front-desk inquiries 62% and time-to-answer 68% in 8 weeks.
HIPAA-aware self-serve booking with two-way reminders lifted kept appointments 27% and cut no-shows 29% across 6 dental locations.
These are past engagements at the prices those clients paid. For build price, monthly cost and payback side by side, see what ten client projects cost.
AI receptionist, answered
How much does an AI receptionist cost?
A Gigabit AI receptionist is built, integrated and deployed for one flat fee, from $8,000 for a Standard build covering chat and up to two integrations, such as your scheduling system and your website. Voice, three or more integrations and HIPAA scope are priced higher, and you get the exact number in writing after the scoping call. The fee includes 30 days of tuning after launch. After that, Managed AI Operations keeps it reviewed and updated at $3,000–$20,000 a month. Telephony, messaging and model usage are separate running costs that depend on volume.
Can an AI receptionist answer phone calls?
Yes. There are two levels. A voice receptionist answers in natural speech, books and transfers to staff; a phone menu sends routine callers to self-serve by text or the web and puts everyone else through. Our published clinic build used the second, with phone-menu deflection, after-hours replies, web chat, SMS and WhatsApp. We have no published case of a voice receptionist yet, so we scope voice with a pilot on real calls before it takes the main line.
Does it book into our real scheduling system?
That is the point of building one. The receptionist reads live availability and writes bookings, reschedules and cancellations into the system your staff already use, with your rules for visit types, durations and providers. The dental group's booking flow syncs with Dentrix in real time, with conflict handling and rules for high-value services. If a system has no usable connection, we say so at scoping and offer a bridge or a booking request a person confirms.
Is an AI receptionist HIPAA compliant?
No AI receptionist is compliant on its own; the whole system has to be. For practices handling patient information we build HIPAA-aware systems: a business associate agreement with every company whose product touches PHI, Gigabit included, only the fields each task needs, encryption in transit and at rest, logged conversations and actions, and clinical questions passed to staff rather than answered. We do not claim a HIPAA certification; we list the controls in place. Your privacy officer makes the final call.
Can an AI receptionist work for a dental practice?
Yes, and dental front desks are a good fit: many calls are bookings, reschedules, insurance and preparation questions. Our published dental build gave a six-location group self-serve booking synced with Dentrix and two-way SMS and WhatsApp reminders with links to confirm or reschedule. Within 60 days kept appointments rose 27% and no-shows fell 29%. A conversational receptionist adds answers and booking inside a chat or call on top of that.
Is an AI receptionist the same as an AI appointment setter?
They overlap. An AI appointment setter usually means booking: taking an inbound request and putting it in the calendar, or following up leads to book a meeting. An AI receptionist also answers questions, routes calls and hands over to staff. We build both, and for most service businesses one agent does the inbound part of each.
What happens when the AI receptionist cannot help?
It hands over instead of guessing. Keyword and intent triggers route complaints, urgent and clinical questions and anything outside its approved answers to a person, and the handoff carries a summary so the customer does not repeat themselves. In the clinic build, handoffs to staff carry a conversation summary, and the assistant gives an explicit not-medical-advice framing with refusal rules.
Often scoped alongside.
Gigabit Agents
The full catalog of production agents, including voice and booking, from $8,000 flat.
Gigabit AgentsAI for healthcare
HIPAA-aware intake, front-desk and documentation agents, with BAAs and human review.
AI for healthcareClinic patient assistant
Repetitive front-desk inquiries down 62% and time-to-answer down 68% in 8 weeks.
Clinic patient assistantRAG development
AI that answers from your own documents and data, with citations.
RAG developmentFrom the field notes
Missing calls, or answering the same questions all day?
Tell us which channels customers use, what they ask and which scheduling system you run. We reply with what the receptionist would handle first and the flat fee. The $25,000 AI Transformation Sprint is there if you want the wider front desk mapped first.