Automating WhatsApp Orders for Restaurants: Allergies, Multi-Item Carts, Live Tracking

Food ordering over WhatsApp looks simple from the outside — a menu, a few messages, a confirmation. In practice it's one of the harder conversation flows to automate well, because the questions carry real consequences. "Is this dairy-free?" isn't a nice-to-have answer; getting it wrong is a health issue, not a UX bug. A restaurant automating this flow has to get ingredient-level accuracy right before anything else.

Ingredient questions need a real answer, not a plausible one

A customer asking about allergens is asking a factual question with a factual answer that exists somewhere — a product record, a supplier spec sheet, a menu database. The right approach is a direct lookup against that data: ingredients, allergens, nutritional info, whatever the menu system actually stores. If the information isn't in the system, the honest answer is "I don't have that detail — let me connect you with someone who does," not a guess dressed up in confident language. This is the single most important design decision in a food-ordering assistant, more important than speed or tone.

Carts are conversational, not form-based

Real orders rarely arrive as a clean list. "Two spicy chicken burgers and a large fries" is one message that needs to become a structured cart with two line items, correct quantities, and a running total — without asking the customer to repeat themselves in a rigid format. The system should parse the message as given, build the cart, present a clear summary ("2× Spicy Chicken Burger + 1× Large Fries = $X"), and wait for confirmation before anything moves to the kitchen. Customers add items mid-conversation too — "actually add a drink" needs to update the existing cart, not start a new order.

Delivery is where trust either builds or breaks

Once an order is placed, "where is my food?" is the single most common follow-up message in food delivery — and it's also the one most businesses handle worst, because it means someone checking a system and typing a reply manually, often minutes after the customer asked. An automated flow should answer this instantly by pulling live order status rather than making the customer wait for a human to look it up. That single change — instant status vs. a delayed manual reply — has an outsized effect on repeat orders, because "did they forget about me" is the anxiety driving the question in the first place.

Why this matters more for food than most verticals

Food has the shortest patience window of any vertical selling over WhatsApp. A retail customer will wait a day for a reply about a jacket. A food customer messaging at 8pm on a Friday will order from a competitor in the time it takes a human to notice the message. That makes response latency — not just accuracy — a direct revenue variable. An assistant that answers ingredient questions correctly, builds carts from natural language, and tracks orders in real time removes the three biggest points of friction in a channel where friction has an immediate, measurable cost: the abandoned order.

This applies well beyond any single market — restaurants, cloud kitchens, and food delivery operators handling WhatsApp orders anywhere run into the same three problems in the same order: accuracy on ingredients, cart parsing, and delivery status. Solving those three is most of the automation problem for this vertical.

See how Sellatrix works for this industry →