How much does a typical project cost?

Depends on scope, but for reference: a specific automation from USD $300, an AI agent from USD $800, a complete CRM integration from USD $2,500 and a SaaS from USD $8,000. In the diagnosis I give you a tight range for your specific case.

How long to see results?

The first automations usually run in 1-2 weeks. A basic AI agent in 2-3 weeks. A complete SaaS project between 8-16 weeks. I work in short sprints so you see progress in each delivery.

What if I already have tools? Do you replace or integrate?

I integrate whenever possible. The idea is not for you to change everything, but for what you already have to work better and communicate with each other. I only recommend replacing when a tool is costing you more than it contributes.

Do you work alone or with a team?

I personally lead each project and, depending on size, add developers, designers or data experts from the Oryzo team and my network. You always have a single point of contact: me.

Do I need technical knowledge to maintain it?

No. I deliver everything documented, with simple dashboards and training for your team. If you later prefer continuous support, I offer monthly maintenance plans.

Do you handle projects outside Chile?

Yes. I work with clients throughout LatAm and Spanish-speaking USA. Everything is 100% remote, with meetings via Google Meet or Zoom and asynchronous communication via WhatsApp/Slack.

Automating Reservations and Orders via WhatsApp: A Practical Tutorial for Restaurants

A full restaurant on a Friday night sounds great. Until you realize there are three people answering WhatsApp messages at the same time, table 8 has been waiting for a confirmation for 20 minutes, and someone wrote down a reservation wrong — now two groups are waiting for the same spot.

Manually managing reservations and orders via WhatsApp works up to a point. After that point, it starts costing you customers, time, and money. This tutorial explains how to automate that process in a practical way.

The Problem with Managing Reservations Manually

WhatsApp is the preferred channel for customers in LATAM to contact restaurants. It's fast, direct, and convenient. The problem is that the customer's convenience becomes operational burden for your team.

The most common issues:

Uncovered hours: customers write at 11 PM to book for Saturday. If nobody responds, they book somewhere else.
Logging errors: reservations noted on paper, in the server's memory, or in a team WhatsApp group are reservations at risk.
Team time: each reservation message requires a 3 to 6 message conversation to confirm date, time, number of guests, and name. Multiplied by 30 reservations a week, that's hours of work.
No automatic confirmations: customers forget their reservations and there's no way to remind them without manual effort.
Takeout orders: the same problem, multiplied by the urgency of someone who wants to pick up their food in 30 minutes.

The solution isn't hiring more people to answer WhatsApp. It's automating the conversation.

How WhatsApp Reservation Automation Works

The system is based on an AI conversational agent connected to your WhatsApp Business number. When a customer writes, the agent:

Greets them and asks what they need (reserve a table, place a takeout order, check the menu, hours, etc.).
Based on the intent, guides the customer through a specific flow.
Collects all the necessary information (date, time, number of guests, name, contact phone number).
Verifies availability in real time against your reservations system.
Confirms the reservation with the customer and sends an automatic reminder 24 hours before.
Logs everything in a centralized system accessible to your team.

The agent works 24/7, responds in seconds, and handles multiple conversations simultaneously. Your team only steps in for special cases or when the customer requests it.

Step-by-Step Setup

Step 1: Get Access to WhatsApp Business API

This isn't the regular WhatsApp Business app. It's the API, which allows you to connect automation tools. You can access it through providers like:

Meta Business Suite (direct, more technical)
Twilio, 360dialog, or Wati (simpler for non-technical users)

You'll need a dedicated phone number for the restaurant that isn't already using the WhatsApp app.

Step 2: Configure the Automation Engine

We use n8n as the flow engine. This is where the logic is defined: what happens when a customer says "I'd like to reserve," what data is asked for, how availability is checked, and how it's logged.

The basic reservations flow in n8n includes:

Entry node via webhook (incoming WhatsApp message)
Intent classification node (reservation, order, or inquiry?)
Conversation node with the AI model to collect data
Availability check node against the calendar/system
Confirmation and logging node
Scheduled reminder node (24h before)

Step 3: Configure the Availability System

The agent needs to know when tables are available. Options depending on complexity level:

Simple: Google Sheets with the reservations calendar. The agent queries and updates the sheet.
Intermediate: Airtable or Notion as a reservations database.
Advanced: integration with your POS or restaurant management system (like Toast, Square, or similar).

For most mid-sized restaurants, Google Sheets is enough to start.

Step 4: Build the Knowledge Base

The agent needs to know your restaurant in order to answer basic questions without escalation:

Hours of operation
Address and how to get there
Menu (or link to the digital menu)
Reservation policies (minimum advance time, maximum group size, deposits)
Takeout options
Accepted payment methods

This document feeds the AI model and is the foundation for all responses.

Step 5: Testing Before Launch

Before activating the system for real customers:

Run 20–30 test conversations covering every possible scenario.
Test edge cases: what happens if there's no availability? If someone wants a table for 15? If they ask about a specific dish?
Verify that records arrive correctly in the availability system.
Confirm the team knows how to access the system and what to do when there's an escalation.

What Happens When the Agent Doesn't Know the Answer

This is a critical point. A well-designed agent knows when it doesn't know.

When a customer asks something outside the agent's scope — a very specific special request, a complaint, an unusual situation — the system should:

Acknowledge the limit: "This question needs to be answered directly by someone on our team."
Hand off to a human: Notify the team in real time with the conversation context.
Don't leave the customer without a response: Let them know a human will respond within X time.

The worst mistake is an agent that makes up answers or goes in circles without resolving anything. An honest, quick escalation is better than an incorrect response.

Expected Results

Based on similar implementations in restaurants in Mexico and Chile:

Response time: from minutes or hours to under 30 seconds for any inquiry.
Off-hours reservations: restaurants capture between 20 and 35% of their reservations during hours when there was previously no human coverage.
Reservation errors: practically eliminated with a centralized and automated system.
Team workload: 60–70% reduction in messages requiring a manual response.
No-shows: 25–40% reduction thanks to automatic reminders.

The impact on sales is direct: fewer empty tables from lost or mismanaged reservations, and more customers served with the same team.

Is Your Restaurant Ready to Automate?

If you manage more than 20 reservations per week via WhatsApp, automation already makes financial sense. If you have more than 50, it's urgent.

I'm Jasiel Tellez, an automation specialist for restaurants and small businesses in LATAM. I've implemented automatic reservation systems for restaurants that now operate with less stress and more customers served.

Schedule a free diagnostic call →

In 30 minutes we'll review your current operation and I'll present a concrete proposal tailored to your restaurant.