Home » Use Cases

Automate Your Food Ordering Process

High-level overview of food ordering on Answers

Today, many customers want their brand to be where they are – on their favorite social media or chat app. In order to be there for your customer, you can build and deploy a chatbot on their preferred chat app and enable offer them with food ordering directly from that app.
That way, you can open up a new delivery stream and at the same time directly engage with your customers. A chatbot can receive and make orders, and send delivery notifications, or even promote your special offers.

If you would like to set up a chatbot for food delivery, read steps below.

This use case will demonstrate how to define, design, build, and deploy a chatbot which can help your customers order food while chatting through an app.

PROCESS WORKFLOW

Process workflow for food ordering bot in Answers

IMPLEMENTATIONS STEPS

Pre-requisites:

  1. Before you design your chatbot, sort your conversations into different categories (ordering, tracking, menu, etc.), based on what you would like to communicate through a chatbot.

  2. Based on the categories, think about what topics fall under each of them (what types of food can be ordered, what are different delivery options, etc).

  3. Create a schema for your conversations and based on it, you can now design the dialogs for the chatbot.

Steps over Web Interface

  1. Log in to the web interface, navigate to the Answers module > NEW CHATBOT. Refer to the Create Chatbot article.

  2. The Bot name and Channel are mandatory fields. Select the channel you most commonly use to communicate with your customers. For example, WhatsApp. Before you activate the chatbot, make sure you also configure the rest of settings, such as the Sender number, Language, Escape phrases, and Session Timeout.

  3. Based on the pre-requisite steps, create the required attributes. Use Attributes to save useful information about your customer, like name and surname, phone number, food preferences, delivery address, etc.

  4. Since the conversations will branch according to keywords (chatbot will use them to route your customer to the correct dialog), create keywords. Enter as many synonyms as you identified are possible variants of the keywords (e.g., the order keyword can have synonyms like 1, orders, pizza, or similar).

  5. Once you've created attributes and keywords in the Answers platform, start designing your dialog. Based on the conversation schema, use the necessary elements to build the dialogs.

NOTE

As this is a keyword-based bot, you do not need to select the intent for the initial element of your dialogs.

  1. You can use the Default dialog for the initial message you will send to your customers as all messages that don’t have a specified dialog entry point will end up here.

Example of keyword-based dialog for food ordering chatbot

  1. Based on the keywords you defined in the previous step your customer can enter 1, order, or pizza and the bot will know to proceed with the pizza order.

  2. Use the Process User Input element to branch different keywords into further dialogs.

Conditions dialog for food ordering

  1. Use elements like Attribute, Conditions or Send text to create dialogs according to your design. These dialogs will enable the chatbot chat with customers regarding the food they want to order, where they want it delivered, how they want to be charged, etc.

  2. When you save attributes, use those values to confirm the specifics of the order, like the type of a pizza, number of pizzas, or the delivery address.

  3. Once all the dialogs are ready, use the Simulator to test how well they perform in simulations of real-life conversations.

Simulation of chatbot and end user conversation of ordering pizza

  1. When the simulations pass according to your design, activate the bot.
PRO TIP

Use message randomization to vary the messages sent to your customers and make the experience of ordering food different and exciting every time each time it happens.

ADVANCED INFO

In some situations, a chatbot is going to be unable to resolve your end users’ issues. In order to keep them satisfied, use “Redirect to Agent” element and transfer the conversations to a live-agent, by using Conversations, our cloud contact center solution.