Boni
Back to case studies

Food distribution · Malaysia

AI ordering for a multi-outlet Malaysian food group

A publicly listed Malaysian food and poultry group turns WhatsApp messages, voice notes, and product shorthand into clean orders—without an experienced employee decoding every single one by hand.

Customer workflow

Five outlets first. Built for the whole group.

Product mix

Bow ChatBoni AI AgentsOrder Operations

Challenge

Orders don't arrive as orders. They arrive as "the usual, 2 boxes" at 11pm. As a voice note in mixed languages. As a product nickname exactly one person in the company understands. Someone experienced has to decode every one of them—and that person becomes the bottleneck for the whole outlet. The group needed a dependable order record without stopping to build an ordering API first.

Inside the customer operation

A customer sends a voice note asking for the usual product in two boxes, then adds a second item using a local nickname. The outlet needs a draft with the right customer, product, quantity and unit before staff confirm the order.

How the work moves

Resolve the customer and outlet

The conversation is matched to the customer and the outlet that serves them. Catalog choices and history belong to that relationship, so one outlet’s shorthand does not change another outlet’s order.

Read the request in context

Text and voice-note content are interpreted alongside the customer’s prior orders. Product aliases and units of measure narrow the selection; incomplete requests go back for clarification.

Keep a reviewable draft

Staff see the matched products, quantities and relevant conversation before confirmation. Ambiguity enters review instead of silently becoming a committed order.

Continue after confirmation

Order status, changes, escalations and the daily reporting export remain connected to the conversation. The next shift sees the order and the reason for a review decision.

The customer operation

Order intake runs across the first five outlets—and the group never had to build an ordering API to get there.
Staff confirm drafted orders instead of reconstructing shorthand, voice notes, and repeat orders by hand.
The order model already fits the group's wider outlet network, so the next outlets switch on without a new implementation.

What to measure in your operation

  • Review order completeness and product/unit corrections.
  • Track clarification, confirmation and unresolved-order queues.
  • Compare time from enquiry to staff-confirmed order using the outlet’s own workload.

Explore the product and setup

Next Step

Map this pattern to your business.

Talk to Boni