Salon Copilot AI

An AI employee for small salons that automates appointments with LINE, GPT-5.6, Codex, n8n, and Google Calendar.

Comment

Inspiration

I currently work as a hairstylist in a salon.

Salon work involves much more than providing services and interacting with clients. It also includes checking appointments, handling schedule changes, updating calendars, and replying to customer messages.

When an appointment request arrives through LINE during a service, I cannot always respond immediately. I often have to check messages once I am free, which can leave customers waiting or allow their preferred time slot to become unavailable.

I want to focus on the client in front of me, but appointment requests cannot simply be ignored. Constantly switching between salon work and administrative tasks had become a significant part of my daily workload.

This is especially difficult for independent and small salons, where hiring a dedicated receptionist is often not realistic. Delayed replies and missed requests can directly lead to lost booking opportunities.

I wanted to create a system that would reduce the time hairstylists spend managing appointments and allow them to focus more fully on their clients and their work.

That idea led me to develop Salon Copilot AI, an AI receptionist designed to support day-to-day front desk operations in salons.

What it does

Salon Copilot AI is an appointment management system that uses AI to understand booking and rescheduling requests sent through LINE.

Appointment requests through LINE

Customers can send messages in the same way they normally use LINE.

For example:

“I’d like to book a haircut tomorrow at 3:00 PM.”

“Do you have any availability next Thursday afternoon?”

The AI identifies the requested date and time, service, and customer information from the message.

Availability checks and alternative time suggestions

The system checks the salon’s schedule in Google Calendar to determine whether the requested time is available.

If the time slot is already booked, it suggests other available options.

Final approval by the salon owner

The appointment details identified by the AI are displayed in a dedicated dashboard.

The appointment is added to Google Calendar and the customer is notified through LINE only after the salon owner reviews and approves it.

Because the AI does not confirm appointments on its own, the risk of incorrect or unintended bookings is reduced.

Customer information management

Customer information collected during the booking process is saved to Google Sheets.

This makes it possible to view customer details and appointment history in one place.

Google review reply assistance

The AI creates a draft response based on the content of each review.

Salon staff can review and edit the response before using it.

How we built it

We used n8n to manage the project’s workflows.

At the center of the system, n8n connects the LINE Messaging API, Google Calendar, Google Sheets, and GPT-5.6.

When a message is received through LINE, the AI analyzes the text.

It then separates the message into the information required for the appointment, including the date, start time, end time, service, and customer details, and sends that information to the dashboard.

Once the salon owner approves the request, the appointment is added to Google Calendar and the customer receives a confirmation message through LINE.

GPT-5.6 was used to support the overall system design, booking flow, message interpretation, and prompt design.

Codex was used for dashboard development, code generation, debugging, error investigation, and workflow improvements.

AI was not only used as part of the final product. It also supported the development process by helping with implementation and problem-solving.

Whenever something did not work, I investigated the cause, made changes, and tested the workflow again.

Challenges

One of the most difficult parts of the project was converting vague LINE messages into structured appointment data.

For example:

“Next week in the evening.”

“The same time as before.”

“A little later.”

These expressions are natural in conversation, but Google Calendar requires a specific date, start time, and end time.

I therefore had to build a process that extracts the necessary information from each message and converts it into a format that can be used to create a calendar event.

Rescheduling and alternative time suggestions created an additional challenge because the system needed to retain information from earlier messages.

Since a conversation can continue across multiple messages, the system also had to keep track of which appointment the customer was referring to.

I repeatedly adjusted the conversation state, conditional logic, prompts, data transformation, and error handling in n8n.

Another challenge was deciding how much of the process should be automated.

If everything were handled automatically, a misunderstanding by the AI could result in an incorrect booking.

To address this, I added a final review step in the dashboard.

This approval flow allows the system to automate part of the appointment process while keeping the final decision in the hands of the salon.

What's next

In the future, I plan to expand Salon Copilot AI beyond appointment management and support a wider range of daily salon operations.

Appointment reminders

The system will automatically send confirmation messages to customers the day before or on the day of their appointment.

I also plan to add a flow that can handle appointment confirmations and rescheduling requests based on the customer’s reply.

Prepayments and deposits

The system will connect with payment services such as Stripe to support prepayments and booking deposits.

The goal is to reduce losses caused by no-shows and last-minute cancellations.

Customer data analysis

The system will analyze appointment history, services received, and visit frequency.

I plan to expand the dashboard so salon owners can identify customers who have not returned recently and customers who may be likely to book again.

Multilingual support

The system will also support messages from international customers.

The goal is to reduce the workload created by language differences and make it easier for salon staff to respond.

Through Salon Copilot AI, I want to reduce the time hairstylists spend managing appointments and create an environment where they can focus on the customer in front of them.

Built With

Share this project:

Updates