Key Results
Introduction
Over 20 minutes of WhatsApp back-and-forth. That was the cost of every booking at Bruno's Cabeleireiros. Today it's 3.
Bruno's, a salon in São Paulo, Brazil, brought in Embed Station to fix it. The result is Bella — an AI receptionist who lives inside the salon's WhatsApp, replies in 20 seconds (always), and books appointments straight into the Trinks calendar. The front desk is out of the loop.
"The implementation is just starting, but our customers are happy and impressed with the AI's responses." — Karina, Founder & Owner, Bruno's Cabeleireiros
The Challenge
Before Bella, here's how it went: a message comes in on WhatsApp. Someone at the front desk reads it — when they can. Could be right away, could be half an hour later, because there's a customer in the chair, the phone is ringing, lunch break, coffee break. When they finally open Trinks to check availability, they go back to WhatsApp and ask if the customer prefers Tuesday or Thursday. Wait for a reply. Confirm. Over 20 minutes. Per booking.
And every reply took more than a minute — when it came at all. Some never got an answer. Some gave up entirely.
Booking Time Reduction
Our Approach
Bella isn't a chatbot firing off canned replies. Every time a customer sends a message, she actually checks the Trinks calendar in real time — she sees open slots, which stylists are working that day, what services are offered, and whether the stylist the customer asked for has time. Then she replies with an option that exists. If the customer confirms, she books it herself.
Behind the scenes, Embed Station combined workflow automation, conversation memory, and an AI language model to figure out what the customer wants. The core piece is the integration with the Trinks API — that's where the 20 seconds come from.
Tech Stack
You don't need to understand the stack — Embed Station handles that. For transparency, here's what runs underneath:
Connect to Trinks
Bella sees the same calendar the team sees, in real time.
Build the AI flow
The automation decides what Bella does on every message.
Test until it holds
Real conversations, weird edge cases, until Bella can handle the daily grind.
The Solution
Before, the team did everything. Now Bella handles the repetitive part and the team handles the part that matters — taking care of people, one by one.
Books from start to finish
customer sends a message, picks the service, picks a time, confirms. Done.
Knows what's actually available right now
open slots, stylists scheduled that day, services offered, and whether the requested stylist has time — all from the real calendar, in real time.
Talks straight to Trinks
whatever Bella books shows up for the team the same second.
Who This Works For
This model fits you if:
- You use Trinks, Vagaro, Booksy, Square Appointments, or any booking system with an API.
- Your customers book through WhatsApp.
- Your team spends too much time answering the same questions and booking by hand.
If that sounds like your salon, Bruno's path is your starting point.
Results and Impact
The numbers speak for themselves. Every reply used to take more than a minute — when it came. Today it's 20 seconds, always. A full booking, from the first "hi" to "you're booked," went from over 20 minutes to 3. And the front desk is back to looking at the people in the salon.
Efficiency Gain
Conclusion
Over 20 minutes turned into 3. Every reply now takes 20 seconds, always. Bruno's team is back to doing what they do best — taking care of customers. Active confirmation calls go live in two weeks.
But here's what's worth taking with you: if your AI doesn't check your real calendar in real time, it's just pretending to know. When a customer confirms a slot that doesn't exist — or gives up because no one replied — it becomes your front desk's problem either way. Most salon chatbots are just canned-reply boxes that can't see the calendar. What made Bella work isn't the language model. It's the integration with Trinks.
Run the math. In a salon with an average ticket of R$100 (≈ $18 USD, original BRL converted at 5.5:1 — exchange rate at the time of the case), just 1 customer giving up per week — because no one replied in time — adds up to R$400/month (≈ $73 USD) in vanishing revenue. Over 12 months, that's R$4,800 (≈ $870 USD) gone because the front desk was busy with something else when the message came in. For US salons with average tickets near $75, those numbers roughly triple — over $2,400/year in revenue lost to slow replies. This number lives in your calendar right now. You just aren't looking at it.
The same principle of plug-into-real-data shows up in our analysis case at Bella Capri — different industry, same lesson.
Next Steps for Bruno's
- Roll out active appointment confirmation calls within two weeks.
- Reach out to past customers when their next appointment is due.
Next Steps for You
30 minutes to map it. A few weeks to have it running. Customer replies landing in 20 seconds from day one.
How many customers gave up booking this week because the front desk didn't reply in time?
If you had to check the calendar to answer, this case study is about you.
A free 30-minute audit. You walk away with (1) where your front desk is losing time on WhatsApp today, (2) an estimate of weekly hours recoverable with AI, and (3) a concrete next step toward having Bella (or its equivalent) at your salon.
If in 30 minutes we don't show you at least 5 hours a week of manual work to take off your team's plate, we can stop right there.
This isn't a pitch — it's a diagnostic. You keep the map either way. And we work with the booking system you already use (Trinks, or any other with an API).
📩 contact@embedstation.com 🌐 embedstation.com