Chatbots for Clinics in the Philippines: What to Check Before You Buy

Quick answer: Before buying a chatbot for your clinic in the Philippines, check how it will handle your actual inquiry process: answering questions, using approved prices, helping someone book and bringing in staff when needed. Ask the supplier to demonstrate those steps using your clinic's rules. In my experience, weak instructions are a common mistake. If you do not plan to customize the bot, I would hold off.

Start with how your clinic handles inquiries today

The first thing I check is the clinic's original workflow: what happens from the first inquiry to a confirmed appointment, before adding a bot.

Write down where inquiries arrive, who answers them, where staff find current information and how appointments are confirmed. Include what happens when someone asks a question the first person cannot answer.

Then look for the specific problem you want help with. Is it answering routine questions after hours? Giving consistent service information? Passing booking requests to the right staff member?

Name the person responsible for each step. If a booking request needs staff confirmation, decide who receives it and what happens outside working hours. Set a response expectation the team can actually meet.

Bring that process to the supplier. It gives both of you something concrete to assess during the demonstration.

Customization means clear instructions and current information

One mistake I see owners make is not prompting the bot well. A prompt is the set of instructions that tells an AI chatbot how to respond.

For a clinic, those instructions need to go beyond being polite and encouraging appointments. They should explain which information the bot may use, how it should answer common questions, when it should ask for clarification and when staff must take over.

For example, require the bot to give an approved service price when it has one. When the price is missing, it should say staff need to confirm it. It should never borrow a price from a similar service.

Revflow's review notes document both missing-information problems and conflicting-instruction problems. One pattern is a bot directing someone elsewhere for a price even though the correct answer is already available. Another is outdated information remaining in a reference document.

That is why I would ask a supplier to check the instructions and the information behind them together. A longer prompt alone is no proof that the bot will handle your clinic's questions correctly.

Check prices, policies and promotions together

Ask which document is the approved source for service names, prices, consultation fees, opening hours and active promotions. Also ask who updates it.

Our review notes include a pattern where the bot, automated messages and staff wording disagree about a clinic policy. Fixing the bot's answer alone leaves the other messages unchanged.

Before launch, compare the information used in the chatbot with your staff's saved replies and automated follow-ups. Resolve disagreements with the clinic owner or the person authorized to set that policy.

For promotions, agree on who approves the wording and who removes it when the offer ends. Ask the supplier to demonstrate an update, including how they check that an older answer is no longer being used.

Make booking and staff handover visible

A clinic appointment chatbot should have a clearly defined role in booking.

Ask whether it sends a booking link, collects a preferred date for staff to review, or creates an appointment through a connected booking system. Have the supplier show the result at the other end.

A preferred date is a request. A confirmed appointment needs confirmation from the clinic's booking process. Require the chatbot's wording to reflect that difference.

Revflow's review notes record availability errors where a bot gave a specific date without access to the doctor's schedule. In that situation, require it to collect a preference and pass it to staff without claiming the slot is available.

Also test staff handover. Show who receives the request, how they find the conversation and how they take control. Decide what the bot should say while staff are unavailable.

Keep diagnosis and treatment decisions with qualified clinic professionals. Define which questions must go to them before the bot starts answering inquiries.

Use this checklist during the demonstration

Ask for a test using fictional scenarios and clinic-approved information. Do not use real patient records or conversation screenshots.

Use the languages your clinic needs, including English, Filipino or Taglish where relevant. Change the wording between tests so you can see whether the bot handles the meaning.

Test	What you should ask the supplier to show
A question about an approved service price	The correct service and price, followed by a relevant next step.
A question with no approved answer	A clear limit and a route to staff, without a guessed answer.
A request for a particular appointment time	The correct booking step and an accurate statement of whether it is confirmed.
A request to speak with staff	A visible handover that reaches the assigned person.
A correction or expression of confusion	A response that checks the issue rather than repeating an uncertain answer.
A promotion that has ended	The expired offer is no longer presented as current.

For each test, look beyond the reply. Check the booking record or staff notification where one is expected. Ask what happens if that step fails and who investigates.

Ask how information will be handled

Under the Philippine Data Privacy Act of 2012, information about an individual's health is sensitive personal information.

Before connecting a chatbot to your inbox, ask the supplier to explain:

• What information will be collected, and why?
• Where will conversations be stored, and who can access them?
• Which other providers will receive the information?
• Will conversations be used to train AI models?
• How long will information be kept, and how will deletion be handled?
• What information and permissions are needed before collection begins?

Have the person responsible for your clinic's privacy requirements review those answers. Start demonstrations with fictional information, and agree on what the bot should collect before launch.

Compare the cost of keeping it accurate

Ask for a written breakdown of setup, recurring charges, usage limits, connected services and support. Confirm whether customization, information updates and fixes are included or charged separately.

Give the supplier a practical change to price: a revised consultation policy or a promotion ending. Ask who makes the change, how long it takes and how the result is checked.

Name someone at the clinic to approve information and someone on the supplier's side to maintain it. Agree on how errors will be reported and who can pause the bot while a serious problem is checked.

Treat maintenance as part of the buying decision.

When I would advise waiting

If you do not plan to customize the chatbot, I would hold off. That is my clearest condition for deciding against buying one.

Before paying, make sure someone will approve the clinic's information, help test the replies and handle requests passed to staff. If those responsibilities are still unclear, settle them first.

You should leave a demonstration able to explain what the bot will answer, how booking works, when staff take over and who keeps it current.

For the next step, see Revflow's clinic chatbot page. Bring your current inquiry process and this checklist to the discussion.

Prepared with AI assistance. Migo reviewed and approved the article, and Revflow checked its factual claims before publication.