GoldenrozBook a call

Healthcare chatbot development for scheduling, intake and billing questions

A healthcare chatbot handles the administrative side of patient contact: scheduling, intake forms, practice FAQs and billing questions, on your website, patient portal or messaging channels. Goldenroz builds it with the technical safeguards HIPAA calls for and a handoff to staff, and it does not diagnose or triage. Projects start at $25,000, fixed price per phase.

From $25,000, fixed price per phase

Updated 2026-11-14. By Shai Goldenberg

What usually goes wrong

  • Front desk phones at capacity
    Calls about hours, directions, accepted insurance and appointment changes fill the lines. Patients with urgent needs wait on hold behind them.
  • Paper intake on arrival
    Patients fill in clipboards in the waiting room and staff retype the answers. Visits start late and errors reach the record.
  • A general chatbot near patient data
    Someone on the team is testing a consumer AI tool with real patient messages, with no business associate agreement and no audit trail.
  • Symptom questions in an admin channel
    Patients type clinical questions into whatever box is on the screen. A bot with no rule for this either answers badly or ignores a possible emergency.

What is included

  • Scope and refusal rules
    A written list of what the bot handles and what it declines. Clinical questions get a fixed response that directs the patient to the practice or to emergency services.
  • Scheduling
    Booking, rescheduling and cancelling against your scheduling system through the route your vendor allows. The dedicated appointment booking bot page covers this flow in detail.
  • Digital intake
    Demographics, insurance details and consent forms collected before the visit in a secure session and written to your system, not left in a chat thread.
  • FAQ and billing answers
    Practice information from approved content, and statement or balance questions answered from the billing system after the patient is verified.
  • Technical safeguards
    Per-user access, encryption in transit and at rest, audit logs, data minimization and model access through a provider account covered by your BAA.
  • Handoff and fallbacks
    Transfer to staff with the transcript and fixed replies when a model fails.

What a medical chatbot should not do

The term medical chatbot covers two very different products. One answers questions about symptoms, suggests conditions or decides how urgently someone should be seen. Software of that kind can fall under FDA rules for medical devices and needs clinical validation. We do not build it.

The other kind works on the paperwork around care, and that is our scope. The line is enforced in the build. The bot recognizes clinical content, does not interpret it and replies with a fixed message approved by your clinical lead: contact the practice, or call emergency services if it may be an emergency.

Administrative automation beyond chat, such as prior authorization and denial appeals, is described under AI for healthcare.

HIPAA and where the conversation happens

HIPAA's Security Rule asks for access controls, audit controls, integrity protection and transmission security for electronic protected health information. The Privacy Rule adds the minimum-necessary standard.

Channel choice matters most. A chat inside your patient portal, behind a login, is the simplest to defend. Public messaging apps and text messages are convenient but may not come with a business associate agreement, so ask each vendor before any PHI goes through its service. Where no agreement is available, the bot uses that channel for non-sensitive steps only and moves the patient to a secure session for the rest.

Every vendor in the path that handles PHI needs a BAA with you: the cloud host, the model provider and the messaging provider. We build and document the controls and develop on synthetic or de-identified data where possible. Goldenroz is an engineering firm, not a compliance auditor or a law firm. Your privacy officer and counsel approve the design. A security audit can review an existing patient-facing app first.

Connecting to the EHR and practice systems

Integration runs through what your vendors permit: FHIR APIs, HL7 interfaces, a vendor's scheduling API or your integration engine. Approval for that access often takes longer than the build, so it is the first item in scoping.

Launch is narrow by design: one location or one specialty, read-only where possible, with staff reviewing transcripts daily. Process and timelines follow the parent AI chatbot development service.

How the work runs

  1. Scoping call

    Before the project

    We review real transcripts, pick the intents worth automating and agree what the bot must never do. You get a fixed price.

  2. Prototype

    About 1 to 2 weeks

    A working bot on the top few intents, tested against your past conversations in a private channel.

  3. Production build

    Roughly 4 to 8 weeks

    System integrations, fallbacks, handoff, logging and channel setup. The range depends on integrations and on WhatsApp account approval.

  4. Limited launch

    At launch

    The bot handles a share of conversations or selected hours first. Your team reviews transcripts daily.

  5. Widen

    After launch

    New intents are added in order of volume, each tested against the conversation set.

What it costs

Each phase has a fixed price agreed before work starts. Model usage and WhatsApp messaging are billed by the providers to your own accounts.

See full pricing

  • Support or ordering bot
    Fixed price per phase
    From $25,000
  • Multi-channel or multi-brand platforms
    Scoped per project
    $100,000 and up
  • Scoping call
    Ends with a written scope and a price
    $0
  • Model and WhatsApp usage
    Billed to your own accounts
    Paid to the providers

Who it is not for

  • Symptom checkers, triage or treatment advice. We do not build clinical decision tools.
  • Solo practices with budgets under $25,000. The messaging features in your EHR or patient portal are the better first step.
  • Mental health or crisis support bots.

Who does the work

CTO and technology partner

Shai leads the technology side: business software, SaaS and mobile apps, payments and escrow systems, Postgres security, AI agents built with Claude Code and MCP, and IT and infrastructure projects with senior infrastructure engineers.

Questions buyers ask

How much does a healthcare chatbot cost?

Projects start at $25,000 at a fixed price per phase, set after a free one-hour scoping call. Model usage, hosting and messaging fees are billed to your own accounts.

Is the chatbot HIPAA compliant?

Compliance belongs to the covered entity and its business associates, not to software. We build the technical safeguards the Security Rule describes and document them for your risk analysis.

Can patients use it on WhatsApp or text?

For non-sensitive steps such as hours, directions and a link to a secure session, yes. Whether PHI may pass through a given channel depends on the vendor's terms and a BAA, and that is a decision for your counsel.

What does the bot do with a symptom question?

It does not answer it. It sends the response your clinical lead approved, which directs the patient to the practice or to emergency services, and flags the conversation for staff.

Which model do you use, and does it see patient data?

Claude is our default, with a second model as fallback, accessed through a provider account where your BAA applies. The bot sends the model only the data needed for the request, and every call is logged.

Guides on this topic

Talk to the person who will do the work

One hour, no sales team. You talk to Shai Goldenberg, the CTO, about your systems or your product, and you leave with a clear next step and a price range.

Or send the details

Book a 1-hour consultationWhatsApp us