Can it answer product questions and collect orders through WhatsApp?

It answers from approved business information and collects items, quantities, the address and payment method. Order, stock and payment-status confirmation depend on your system, enabled integrations or team follow-up.

Does it handle the Arabic dialects customers actually use, and voice and image messages?

Yes. It reads whatever dialect shoppers actually type — alongside MSA — and takes text, voice notes, and images, so a customer can send a photo of the product they want and ask in their own dialect. It replies in the register the customer used, not a generic formal script.

How does handoff to a human agent work?

When a shopper raises something the agent should not settle alone — a complaint, a refund, or a custom order — it escalates to a human on your team with the full conversation and customer record attached, so your staff continues with context instead of starting over. Human escalation is a built-in control, and anything specific to one account or order is routed to your team rather than guessed.

Retail support in Arabic, and the order captured with it.

ChatSheba helps answer product and pricing questions and collect request details, including enquiries received after hours. Your team follows cases that need a decision or action.

The problem in retail operations

Product, price and delivery enquiries arrive across channels, including during busy periods or after hours. The team needs a clear path for collecting and following enquiries and their details.

Where a retail order goes missing

A price or stock question from an ad, arriving after the shop has closed.
Nobody rebuilds an order from scattered messages later, because items, quantity, address and payment are taken as fields inside the chat.
Complaints and special requests, handed to staff with the conversation.

What the AI handles

Answers product, price, and availability questions in the brand's native voice.
Takes orders through structured intake — items, quantity, address, payment method.
Captures customer details and preferences into a structured contact profile.
Receives enquiries from ads directing customers to WhatsApp and captures their information for after-hours follow-up.
Escalates complaints and custom requests to human staff with full context.

How Sheba fits

ChatSheba runs the customer-facing agent across WhatsApp Business API, Telegram, web, and a custom app SDK, with brand-voice tuning and media support; custom engineering connects it to your catalog and order systems.

Selling inside the conversation: ChatSheba
In your brand’s voice, it answers product questions and takes orders on WhatsApp, Telegram, the web and your app.
See ChatSheba
Agents for what happens after the order
The exceptions stay with your team. Checking an order’s status in the order system and telling the customer is work an agent can do.
See how AI agents work
Custom links to catalogue and orders
Sheba builds the connection for your business, to the standards of its products, wherever orders need the catalogue and the order system.
Explore custom AI engineering

Behind each order

ChatSheba on WhatsApp, Telegram, the web and your app.
An AI agent that writes to the contact record.
Brand-voice tuning and media support.
Custom engineering for catalogue and order systems.

Integration and data boundaries

Per-deployment data boundaries, brand-voice tuning, and custom contact attributes for last order, preferred branch, and delivery address. Human escalation is built in. The agent speaks in your brand's voice, not a generic one.

PDPL-aware by design
On-prem and regulated-residency paths
Per-deployment data boundaries
Human escalation as a control

Security and data boundaries

Composite industry example

Shoppers ask what is in stock and where their order has got to. This composite comes from our deployments in Yemen: ChatSheba answers from approved product information and collects request and customer details for follow-up. The team follows requests and complaints, while availability and order-status updates depend on the information and integrations available.

Connecting inventory, store operations and customer service

How this pattern is set up

Channels: WhatsApp, Telegram, web, and an app SDK
Systems it reads and writes: the product catalog and the order system
Escalates to: human staff on complaints and custom requests

Part of 30+ production deployments across 14+ industries.

See the deployment pattern

FAQ

Common questions

Talk to a Sheba specialist about retail.

Discuss your use case