Instagram DM and WhatsApp automation for e-commerce
Responding without missing the "is it in stock, when does it ship" questions coming from Instagram DMs, comments, and WhatsApp in online retail: the field setup of e-commerce channel automation with inTusell.

The Instagram inbox of an online store: a new collection post has taken off, and hundreds of comments have dropped under it — "price?", "is it in stock?", "when does shipping arrive?". At the same time five people are writing on WhatsApp, one of them from abroad in English. A one-person team can't get back to all of them within two minutes, and an "is it in stock" question answered late is a lost cart. This article explains how to set up e-commerce Instagram DM automation in inTusell, along with WhatsApp and comment responses. The steps are concrete and need no technical knowledge.
This article is the channel leg of the e-commerce section of our sector-by-sector "how to train your AI" series. You'll find the AI-training side in the main article of the series, e-commerce AI training, and the day-to-day operations in how to use inTusell.
Quick answer
Online stores connect Instagram DMs, Instagram comments, and WhatsApp to inTusell and collect them in a single inbox. WhatsApp connects in a minute via QR and requires no Meta Business approval; Instagram DM connects with one-click OAuth. The AI responds 24/7 in the language the customer writes in. It gives stock, shipping, and product information from the knowledge base, makes up nothing it doesn't know, and hands the conversation to the live team when needed. This way the purchase intent coming from posts doesn't go unanswered.
The channel reality of e-commerce
Questions don't come from a single channel. A new post takes off and comments explode; someone else who sees the same post writes on WhatsApp; a third reaches you from the chat bubble on your website. Outside a campaign, you can manage this spread. During a sale week, you can't.
inTusell collects all of these channels in a single inbox and connects each one to a fully autonomous AI sales assistant. The emphasis here matters: this is not a chatbot, but an assistant trained to behave like an experienced store advisor. It understands the customer's intent, suggests the right product, and routes them to the order and cart. But it never makes up a stock or shipping detail it doesn't know; it leaves that to the team.
Who is it for?
This article is for online sellers who can't fit the speed and volume of incoming questions into a single channel. Typical profiles:
- Boutiques and branded stores that get dozens of product and order questions a day from Instagram and WhatsApp.
- Fashion, cosmetics, and accessory brands that get a flood of comments on new-collection and campaign posts.
- Stores that sell abroad and lose carts because they can't respond in several languages.
- Everyone from a one-person store to a multi-channel brand; the more channels there are, the greater the benefit.
Connecting WhatsApp: QR (no Meta approval) vs Cloud API
WhatsApp is the busiest channel for e-commerce, and there are two ways to connect it. For most stores, the QR method is enough.
| Method | Setup | Meta Business approval | Who it suits |
|---|---|---|---|
| Connect via QR | ~1 minute by scanning a QR from your phone | Not required | Most stores that want to use their existing number |
| Official Cloud API | App setup on the Meta side | Required | High-volume brands that want the official API guarantee |
The QR method works like connecting WhatsApp Web: you scan the square that appears in the panel with the WhatsApp on your phone, and the number stays the same. Nothing changes on the customer's side; they keep writing to the same number. You don't need to change your number or get a new line.
We support the official Cloud API as an option for high-volume brands that want an enterprise guarantee. Choosing between the two takes a single click, and you can switch later.
Instagram DM automation
Connecting Instagram is one-click OAuth: you click "Connect Instagram" in the panel, approve from Meta's permission screen, and the channel is active. There is no manual token copying.
After connecting, DM automation is controlled by a separate gate (instagram_dm_enabled). When this toggle is on, the AI replies to incoming direct messages. If you keep it off, messages still land in your single inbox but the AI doesn't respond automatically; the live team replies. Splitting the gates lets you move gradually — watch DMs for a week first, then turn it fully on. In e-commerce, DMs are where "is this still available?" comes most often, and because the channel is approved, it works from your first day.
Instagram comments: a gate separate from DMs and moving from comment to DM
In online retail, post comments are a hidden sales channel. Comments like "price?", "is it in stock?", "is this color available?" pour in under a new-collection post, and most go unanswered. Every unanswered comment is a customer ready to buy.
In inTusell, comment responses are managed by a separate toggle (instagram_comments_enabled). This is independent of the DM gate. You can turn on comment responses without opening DM automation, or the other way around. One important caveat: Instagram comment automation is not live; it is rolled out gradually depending on Meta platform approval. Your first step is WhatsApp and Instagram DM — these are approved and immediately usable. The comment gate comes into play once it's ready.
Once comments are ready, the typical flow looks like this:
- The customer writes "price?" under the post.
- The AI leaves a short public reply on the comment ("We've sent you the details via DM").
- The conversation moves to DMs, where the AI asks about size, color, stock, and shipping and routes the customer to the order.
This way no long price negotiation plays out in public; the sales conversation moves to a private channel and continues in a single trackable context.
A single inbox (web chat, Messenger, Telegram, email)
Instagram and WhatsApp are the busiest, but not the only channels. inTusell collects all of the following channels in the same inbox:
| Channel | Connection method | Typical e-commerce usage |
|---|---|---|
| QR (1 min) or Cloud API | The busiest order and "where's my package" channel | |
| Instagram DM | One-click OAuth | Product questions after posts/stories |
| Instagram comments | One-click OAuth + separate toggle (gradual, subject to Meta approval) | "Is it in stock?" comments under collections |
| Facebook Messenger | One-click OAuth | Facebook campaign and catalog traffic |
| Telegram | Bot connection | Campaign channel and community followers |
| Web chat | Widget on the site | Product page and cart questions |
| Account connection | Order, return, and corporate requests |
What this means for the team: you don't jump between seven different apps. All conversations appear on a single screen, with the same customer history. A customer who starts on Instagram and continues on WhatsApp gets one single context rather than two separate conversations.
24/7 and multilingual responses
A large share of online shopping happens outside working hours. Someone sees a post in the evening and writes at night; a customer abroad messages across a time difference. The team sleeps, but the AI doesn't. An "is this still available?" question that comes in at night doesn't wait until morning — the AI routes the customer to the order right away. A question left unanswered after hours becomes a product bought from another store before morning.
Multilingual responses are decisive for stores that sell abroad. The AI replies in whatever language the customer writes in: English to someone writing in English, Arabic to someone writing in Arabic. You don't need a separate channel, a separate account, or a translator for this. One setup covers every language.
Re-engaging the silent cart
The quiet payoff of channel automation in e-commerce is winning back the customer who shows interest and then goes quiet. Every store has one: the customer who says "let me think about it" and disappears, or who looks at the cart and stops writing. The inTusell CRM side follows these customers and, when the moment is right, sends a timely re-offer to suitable WhatsApp customers in follow-up. Let's be clear about what that is: not a managed "waiting list," but a polite, well-timed reminder to a customer who has shown interest and is in follow-up. The automating order and product questions article covers the end-to-end automation of order and product questions.
Channel setup strategy
Turning automation on before a campaign period starts is the right move. Here is what we do in the field:
- Prepare the knowledge base first. Define product descriptions, size and color options, shipping and return terms, and working hours in the panel. You can upload PDF, Excel, CSV, a web URL, or free text to the knowledge base. The e-commerce AI training article covers the details.
- Start with WhatsApp and Instagram DM.
instagram_dm_enabledand WhatsApp are approved and immediately usable, and most stores handle their order traffic here. - Wait for comment automation.
instagram_comments_enabledis a separate gate and is rolled out gradually depending on Meta approval. When it's ready, watch the AI's tone on low-risk comments. - Set the operating mode. Use hybrid mode during a busy period: the AI handles the volume, and the team only looks at sensitive conversations like return disputes.
- Scan the label queue once a day. As you approve and correct the AI's responses, the assistant learns the patterns specific to your store.
This flow lets the system settle in before the rush arrives. Instead of firefighting, your team focuses only on conversations that genuinely need a human.
What it isn't
So that nobody puts it in the wrong category, let's be clear:
- It is not a bulk-messaging or spam tool. inTusell is an assistant that responds to incoming questions; it does not send unsolicited bulk DMs.
- It is not a simple chatbot that works by keywords. It is not a "when you type this, say that" scenario tree, but a sales assistant that understands intent.
- It is not a system that changes your number. Your WhatsApp number stays the same, and the customer notices no difference.
- It is not a system that makes up stock, price, and shipping information. It says it does not know and hands off to the team, and it invents nothing that is missing from the knowledge base. This is the most critical point for customer trust.
- It is not a tool that replaces the store. You always offer and ship the product; inTusell is the response and sales layer.
Frequently asked questions
Do I have to change my WhatsApp number?
No. You connect your existing store WhatsApp number by scanning a QR code from your phone; the number stays the same, and the customer notices no difference. For high volume you can optionally use the official Cloud API.
Is Meta Business approval required?
With the QR method, WhatsApp does not require Meta Business approval and connects in a minute. Instagram DM and Facebook Messenger come in with one-click OAuth and are approved. Instagram comment automation, however, is rolled out gradually depending on Meta platform approval.
Does the AI also reply to comments on posts?
A separate toggle manages the comment gate independently of the DM gate, and it is rolled out gradually depending on Meta approval. Your first step is WhatsApp and Instagram DM, which are approved. Once comment replies are ready, the typical flow is to leave a short reply on the comment and move the conversation to DMs.
What happens with a customer writing from abroad or in another language?
The AI replies in whatever language the customer writes in. If you sell abroad, it answers a customer writing in English, Arabic, or German without any separate channel setup.
Does the AI make up stock and shipping status?
No. The AI never makes up a price, stock count, or shipping date it does not know. It says it does not know and hands the conversation to the team. It answers from what is defined in your knowledge base and invents nothing beyond it.
When does the AI hand off to a human?
You choose the operating mode: all AI, all live team, or hybrid. In hybrid mode the AI runs the normal flow and escalates the conversation to the live team when a return dispute, a special request, or a situation outside the knowledge base comes up.
Next step
Connecting the channels is half the job. The real difference comes from training the AI correctly. To set up product information, shipping and return terms, the tone, and the boundaries, start with the main article of the series, e-commerce AI training. Then see the day-to-day operations in how to use inTusell. To go deeper on order and product questions, read automating order and product questions, review the e-commerce solution page, and follow the series via pricing and all posts.
If you'd like to see how it works at your own store, Request a demo or write to hello@intusell.com. We will connect your channels together and watch the assistant handle the first questions live.
You read the blog — now see it live.
Test inTusell live with your own sector scenario in a 20-minute demo.
Related posts
View all
How does an e-commerce brand train the inTusell AI? A step-by-step guide

Automating order, shipping, return and product questions with the knowledge base in e-commerce
