Pakistan has over 150,000 restaurants, and fewer than 3% of them have a working online ordering system. The rest rely on phone calls, walk-ins, and informal WhatsApp conversations where staff manually type menu items, confirm orders, and chase payments. A Lahore BBQ restaurant reported spending 4 hours per day managing WhatsApp orders manually during peak hours — turning away business because the phone was always busy. This lesson builds the complete restaurant order bot that handles menu browsing, cart management, order confirmation, and JazzCash payment — all on WhatsApp, fully automated.

Section 1: Designing the Restaurant Bot Flow

Before writing a single template, map the complete conversation journey a customer will take:

Customer sends "Hi" or "Menu"
         ↓
Bot sends Welcome + Categories menu
(Starters / Main Course / Drinks / Desserts)
         ↓
Customer selects category (e.g., "1" for Starters)
         ↓
Bot sends items in that category with prices in PKR
         ↓
Customer types item number to add to cart
         ↓
Bot confirms: "Added! Your cart: [items + total PKR]"
Customer can: Add more (type number) | View cart ("C") | Checkout ("O")
         ↓
Customer types "O" to order
         ↓
Bot asks for: Name, Address, Phone (if not known)
         ↓
Bot sends order summary + payment options
(JazzCash 0312-XXXXXXX | EasyPaisa 0300-XXXXXXX | Cash on delivery)
         ↓
Customer sends payment screenshot or selects COD
         ↓
Bot confirms order + sends tracking number
         ↓
Restaurant owner gets WhatsApp notification:
"New Order #1234 — [Items] — PKR [Total] — [Address]"

Section 2: Building with WATI

WATI is the most popular WhatsApp automation platform in Pakistan. Here's how to build the restaurant bot:

Step 1 — Set Up the Menu Structure: In WATI's dashboard, go to "Chatbots" → "Create New Chatbot." Create a keyword trigger: when any message contains "menu," "hi," "hello," "order," or "start," fire the welcome flow.

Welcome Message Template:

Assalam o Alaikum! 🍽️ Welcome to *{{restaurant_name}}*

Please choose a category:
1️⃣ Starters
2️⃣ Main Course — Karahi, Biryani, Daal
3️⃣ Grills & BBQ
4️⃣ Drinks & Fresh Juice
5️⃣ Desserts

Reply with the number to browse. Type *HELP* anytime.

Step 2 — Category → Items Flow: For each category, create a reply flow. When customer replies "2" (Main Course):

*Main Course Menu*

1. Mutton Karahi (serves 2-3) — PKR 1,800
2. Chicken Karahi (serves 2-3) — PKR 1,200
3. Special Biryani (full) — PKR 950
4. Daal Makhni — PKR 450
5. Palak Gosht (serves 2) — PKR 1,100

Type the number to add to cart.
Type *BACK* to see categories.
Type *C* to view cart.

Step 3 — Cart Management: WATI supports session variables. Store the cart as a JSON string in the customer's session:

// WATI webhook or Zapier/n8n handler:
cart = JSON.parse(session.get("cart") || "[]");
cart.push({item: selectedItem, price: itemPrice, qty: 1});
session.set("cart", JSON.stringify(cart));

Step 4 — Payment Integration: When customer checks out, send payment options:

*Order Summary*
{{cart_items_list}}
*Subtotal: PKR {{subtotal}}*
Delivery: PKR 150
*Total: PKR {{total}}*

Payment Options:
💚 *JazzCash:* Send to 0312-XXXXXXX (Account: {{restaurant_name}})
🔵 *EasyPaisa:* Send to 0300-XXXXXXX
💵 *Cash on Delivery:* Available for orders under PKR 3,000

After payment, send screenshot here. For COD, reply *COD*.

Step 5 — Restaurant Notification: Use an n8n webhook or WATI's Zapier integration to send the restaurant owner a WhatsApp notification for every confirmed order with all order details, customer name, and delivery address.

Section 3: Menu Data Management

For easy menu updates without touching the chatbot config, store your menu in Google Sheets. Build an n8n workflow that:

1. Reads the menu from Google Sheets every hour
2. Updates the WATI chatbot templates via WATI's API if any item/price changed
3. Sends you a WhatsApp confirmation: "Menu updated: Mutton Karahi price changed to PKR 1,900"

This means restaurant staff can update prices in Google Sheets, and the WhatsApp bot updates automatically — no technical knowledge required.

Section 4: n8n Workflow Architecture

Google Sheets (Menu DB)
        │
        ▼ (every 1 hour)
┌───────────────────┐
│  n8n Scheduler    │
│  - Read Sheet     │
│  - Compare prices │
└────────┬──────────┘
         │ if changed
         ▼
┌───────────────────┐     ┌──────────────────────┐
│  WATI API call    │────►│  Update bot templates │
│  POST /templates  │     │  (new prices in PKR)  │
└────────┬──────────┘     └──────────────────────┘
         │
         ▼
┌───────────────────┐
│  WhatsApp notify  │
│  to restaurant    │
│  owner            │
└───────────────────┘

Customer Order Flow (real-time):
Customer → WATI Bot → n8n webhook → Google Sheets (order log)
                                  → WhatsApp notify to owner
                                  → Auto-reply with order #

Section 5: Sample n8n Order Processing Node

# Python equivalent of n8n's function node for order processing
import json
from datetime import datetime

def process_order(webhook_data: dict) -> dict:
    """Process incoming WhatsApp order from WATI webhook"""

    customer_phone = webhook_data["contact"]["whatsapp_number"]
    cart_json = webhook_data["session_variables"].get("cart", "[]")
    cart = json.loads(cart_json)

    # Calculate totals
    subtotal = sum(item["price"] * item["qty"] for item in cart)
    delivery_fee = 150 if subtotal < 3000 else 0  # Free delivery over PKR 3,000
    total = subtotal + delivery_fee

    # Build order record
    order = {
        "order_id": f"ORD-{datetime.now().strftime('%Y%m%d%H%M%S')}",
        "customer_phone": customer_phone,
        "items": cart,
        "subtotal_pkr": subtotal,
        "delivery_pkr": delivery_fee,
        "total_pkr": total,
        "status": "pending_payment",
        "timestamp": datetime.now().isoformat()
    }

    return order

# Expected output:
# {
#   "order_id": "ORD-20260401143022",
#   "customer_phone": "+923001234567",
#   "items": [{"item": "Mutton Karahi", "price": 1800, "qty": 1}],
#   "subtotal_pkr": 1800,
#   "delivery_pkr": 150,
#   "total_pkr": 1950,
#   "status": "pending_payment"
# }

Exercise 1: Map a complete order flow for a fictional Pakistani restaurant of your choice (BBQ, Chinese, Desi, Seafood). Create a flow diagram with: welcome → categories → items → cart → checkout → payment → confirmation. Identify every point where the bot needs to wait for customer input versus send an automatic message.

Exercise 2: In WATI's free trial (available at wati.io), build the welcome flow and one complete category flow (at minimum 5 items with PKR prices). Test it by WhatsApp messaging your own test number. Verify the bot responds correctly to: selecting a category, adding an item, viewing the cart, and typing "BACK."

Exercise 3: Build the Google Sheets menu manager. Create a sheet with columns: category, item_name, description, price_pkr, available (TRUE/FALSE). Add 10 items. Write an n8n workflow that reads this sheet and prints the menu in WATI-formatted text. This is the foundation for automated menu management.

Pakistan Case Study: Lahore BBQ House

The problem: Lahore BBQ House (fictional, based on real archetype) received 120+ WhatsApp messages daily during dinner rush (6–9 PM). Two staff members spent 4 hours managing orders. Common errors: wrong items, unconfirmed orders, payment disputes.

The solution: Restaurant order bot with Google Sheets backend.

Setup: 1 week build, WATI Business plan (PKR 18,000/month), Google Sheets (free), n8n (PKR 3,500/month).

Results after 30 days:

• Staff time on order management: 4 hours → 20 minutes (flagging exceptions only)
• Order errors: 15/week → 1/week (bot captures exact items, no misheard orders)
• Average order value: PKR 1,200 → PKR 1,650 (bot shows full menu, upsells naturally)
• Customer satisfaction: Higher — instant response vs 20 min wait
• Monthly revenue increase: PKR 180,000
• Total automation cost: PKR 21,500/month
• Net gain: PKR 158,500/month

Key Takeaways

• The restaurant order bot is Pakistan's highest-ROI WhatsApp automation because it replaces 3-4 hours of daily manual work that every restaurant currently does
• Session variables in WATI allow cart persistence across a multi-message conversation — without them, the bot would forget what the customer already added after each message
• Google Sheets as a menu backend makes the system maintainable by non-technical restaurant staff — a crucial feature for real client deployments
• Always include a COD (Cash on Delivery) option in Pakistani restaurant bots — a significant percentage of customers, especially first-time buyers, will not pay digitally before trying the food
• The n8n notification layer ensures the restaurant owner is always informed of new orders in real time, bridging the automation with human oversight where it matters