Repeated product questions
Customers regularly need product lists, prices, stock information, delivery details, and basic product answers that should not require manual replies every time.
Production Client Case Study
WhatsApp ecommerce automation for product questions and orders.
This WhatsApp ecommerce automation for product questions and orders was built for a confidential retailer that needed a faster way to handle repetitive product questions and turn customer conversations into structured orders while keeping predictable business logic in control.
The Challenge
Why WhatsApp ecommerce automation for product questions and orders needs controlled business logic.
The goal was not to build a chatbot that sends every message to AI. The goal was to automate repeatable customer-service and order steps while keeping the conversation practical and controlled.
Conversation context matters
The system needed to remember selected products and the current conversation stage so customers could continue naturally across multiple messages.
Orders need several handoffs
Order details had to be collected, totals calculated, records saved, the customer confirmed, and the business owner notified without manually copying the same information between tools.
How the System Works
One WhatsApp conversation can move from product discovery to a recorded order.
The workflow uses Meta WhatsApp Cloud API and n8n as the automation layer, with product data, session memory, AI routing, order storage, and notifications connected behind the conversation.
01
Customer sends a WhatsApp message
Meta sends the customer message to the n8n webhook. The workflow normalizes supported message types and filters delivery/read callbacks that should not be processed as customer requests.
02
Business logic handles predictable requests
Greetings, product menus, product selection, price, stock, and other structured questions can be answered from store data without making an unnecessary AI call.
03
AI is used only when reasoning adds value
Recommendation, comparison, and natural-language questions can be routed to OpenAI with controlled product context rather than unrestricted business data.
04
Order information is collected and recorded
When the customer decides to buy, the conversation collects the required order information, calculates the payable amount, and saves the confirmed order in Google Sheets.
05
Customer and business are notified
The customer can receive a WhatsApp confirmation and email while the business owner receives an automated new-order alert.
System Components
The automation connects conversation, product data, AI, and order operations.
Each part has a clear job so AI does not replace predictable business logic unnecessarily.
Product data
Product names, descriptions, prices, stock status, delivery estimates, and notes are available to the workflow for structured answers.
Customer session memory
The workflow can retain the selected product, conversation stage, and order details needed to continue a multi-message interaction.
Smart AI routing
A decision layer determines whether the request needs AI. Straightforward store questions remain deterministic; recommendations and comparisons can use OpenAI.
Orders & notifications
Confirmed orders are stored in Google Sheets, followed by customer confirmation and an automated business-owner alert.
Verified Production Testing
The workflow was tested through real customer-style conversation and order flows.
The case study uses functional outcomes from the implemented system rather than invented conversion percentages or sales claims.
Product conversation worked
A customer greeting displayed the product menu, and selecting a product returned its price, stock, delivery estimate, description, and ordering instructions.
AI routing worked when needed
A recommendation question such as which product was suitable for university use was identified as a reasoning request and routed to AI using the available store data.
End-to-end order flow worked
One successful test created a Wireless Mouse order at Rs. 1,999 plus Rs. 200 delivery, saved the Rs. 2,199 total to the order flow, and triggered WhatsApp, customer-email, and owner notifications.
Production Lessons
Real deployment required more than building the workflow canvas.
The project included Meta webhook verification, WhatsApp Business configuration, API permissions, event subscriptions, and troubleshooting the difference between real customer messages and WhatsApp status callbacks.
Meta integration
Webhook subscriptions, access configuration, and production WhatsApp events had to be connected correctly before incoming messages could reach n8n.
Event filtering
The workflow distinguishes customer messages from delivery and read-status callbacks so irrelevant Meta events do not trigger the customer-response logic.
Testing exposed an edge case
A stale-session edge case was identified during testing, showing why conversation-state handling needs continued validation as the automation evolves.
Technology Stack
Built with connected tools rather than a single chatbot.
The production system combines n8n, Meta WhatsApp Cloud API, OpenAI, Google Sheets, Gmail, JavaScript, webhooks, and HTTP/API integrations.
Questions
About this production WhatsApp automation case study.
The client identity is intentionally anonymized, but the project architecture and functional testing described here come from the real implementation.
Is this a real client project?
Yes. This was built as a production automation system for an e-commerce retailer. The client name and private customer information are withheld from the public case study.
Does every WhatsApp message go to AI?
No. Predictable requests such as product menus, selections, prices, and stock information use business logic and structured data. AI is used mainly when recommendations, comparisons, or natural-language reasoning are useful.
What happens after an order is created?
The implemented flow can save the confirmed order in Google Sheets, send the customer a WhatsApp confirmation and email, and notify the business owner about the new order.
Build a Similar System
Want to turn WhatsApp into a connected business workflow?
CodelyHub can map your product or service data, customer conversation flow, order or lead process, notifications, and AI requirements into a practical automation system.
Best first message
Tell us what customers repeatedly ask on WhatsApp, what information your team handles manually, and what should happen after a lead or order is confirmed.