# Clinic Reply Speed to Appointment Messages: Measure It in 15 Minutes

**When did an appointment message that arrived at night last get an answer the next morning, after the patient had already booked somewhere else?**

> To measure how fast your clinic replies, do what a patient does: send an appointment message at different times, such as midday, night and Friday, on every channel (WhatsApp, Telegram, Instagram DM, phone), and note when the first useful answer arrives. This 15-minute test shows where appointments are being lost.

Author: Saman Ghahreman · Published: 2026-11-21 · Updated: 2026-11-21 · Canonical page: https://solocodelab.com/en/blog/clinic-response-speed-self-audit/

Most clinics do not know how long an appointment message waits for an answer, because nobody has measured it. The receptionist replies between tasks, night messages sit until morning, and DMs get lost. This guide gives a simple test you can run yourself, with no tools and no cost, in 15 minutes.

## Why does reply speed matter for a clinic?

A patient who sends an appointment message has usually decided right then and may have messaged another clinic too; the first clear answer can win the appointment. A well-known Harvard Business Review study on following up sales leads found that replying within the first hour makes a real conversation far more likely than replying later. That study looked at sales companies, not clinics, so treat it as a sign of direction, not a number for your clinic. That is why you should measure your own.

## How do you run a 15-minute reply-speed test?

### Four steps
1. **List your channels**: Every way a patient can ask for an appointment: WhatsApp, Telegram, Instagram DM, phone and the website form.
2. **Send test messages**: From a friend’s number, at three different times (a working midday, a night, a Friday), send a simple message such as “Hello, which day do you have an appointment?”.
3. **Record the time**: Note when you sent it and when the first useful answer arrived. A reply like “thanks, stay tuned” does not count as useful.
4. **Score it**: Use the table below to score each channel and time, then fix the weakest one first.

| Time to first useful answer | What it means | Suggested action |
| --- | --- | --- |
| Under 5 minutes | In good shape | Keep the same speed at other hours |
| 5 to 60 minutes | Acceptable but uneven | Prepare ready answers for repeated questions |
| Several hours | Appointments are probably lost | Add an out-of-hours auto reply and assign someone to follow up |
| Next day or never | A serious problem | Start with a simple auto reply, then consider full automation |

This table is a rule of thumb, not taken from a specific study; adjust the boundaries to your own clinic.

## Three fixes that work within a week

1. A greeting and an out-of-hours message: the WhatsApp Business app has greeting messages, away messages and quick replies built in and is free to start with. Write in the away message when people will get an answer.
2. Ready answers for five repeated questions: opening hours, address, how to book, what to bring and which doctor is in on which day. These cover most messages.
3. A daily unanswered list: one named person, twice a day, reviews the messages that got no reply. You need no system for this; one sheet of paper is enough.

## When should you move to AI?

If after these fixes messages still go unanswered at night and on Fridays, or the receptionist cannot keep up with the volume, automation is worth it: a bot answers repeated questions in seconds and routes appointment requests to a human for follow-up. For how a clinic chatbot differs from a simple auto reply, read [a chatbot for clinics and medical offices](https://solocodelab.com/en/blog/chatbot-for-clinic-and-medical-office/), and for regular follow-up see [automatic customer follow-up with n8n and a CRM](https://solocodelab.com/en/blog/automatic-customer-follow-up-n8n-crm/).

One important red line: a clinic bot should only answer about appointments, opening hours and general information, never give treatment advice, and you should not collect patient medical details in messages. If a patient asks, the bot hands over to the receptionist or the doctor.

> If the test shows messages wait too long, the [automatic AI reply and follow-up service](https://solocodelab.com/en/services/ai-automation/) solves exactly this: you see a live sample first and pay afterwards.

## Frequently asked questions

### How many test messages are enough?

About five to ten messages across three different times (a working midday, a night and a Friday) give a reliable picture. The aim is not an exact number; it is finding your weakest channel and hour.

### Do I have to add AI?

No. First try the greeting message, the away message and ready answers. Move to a bot only if the volume or the unanswered hours still cause trouble.

### Is an automatic reply bad for patients?

If it says clearly that it is a bot, gives only general information and appointment details, and shows the way to a human, it is usually better than hours of silence.

### What happens to patient data?

Do not collect medical details in messages. Let the bot take only a name, a phone number and a preferred time, and leave medical questions to the receptionist or the doctor.

## Sources

- [The Short Life of Online Sales Leads](https://scholarsarchive.byu.edu/facpub/9711) — Harvard Business Review (Oldroyd, McElheran, Elkington, 2011)
- [WhatsApp Business app features](https://www.whatsappbusiness.com/products/business-app-features/) — WhatsApp Business
