Getting started

Building Chatbots

The Web App

Additional resources

Planning your Conversational AI

Anyone can build a bot with Botco.ai — no code necessary. Here's how to prepare and plan for designing your first one.

#1. Choose a goal and map your flow.

Decide on the goal for your bot. Is it to gather and qualify leads? Answer questions? Push people to your online store to make a purchase? Book an appointment?

Choose just one main purpose to start with, and map out a conversation of questions and answer options that will bring users to that desired end point. You can use pen and paper to do this, a digital tool like Miro.com or Draw.io. Our customer success team will usually help you out with this.

Here's an example of a more complex bot flow mapped out in Miro. ⤵️

However, the most successful bots we see normally start with a very simple flow that answers FAQs. A chatbot flow doesn't need to be huge and complex in order to be useful—it's actually quite the opposite! Think about the main problem you're trying to solve and start there. You'll be able to build additional functionalities into your bot as you go.

#2. Plan a bot that’s proactive.

Make your bot a proactive tool instead of a reactive one. Have it ask questions and offer predefined answer options for the user to choose from, in the form of buttons or quick replies.

Having your bot lead the conversation like this will minimize the work the user has to do to get the information or assistance they need. The result? More conversions, and a more streamlined experience!

#3. Plan your bot's persona.

Your bot will act as a virtual ambassador for your brand, so it needs to make the right first impression. Plan out a friendly Welcome Message that lets the user know they’re chatting with a bot, and tells them how it can help. You may even want to choose a name for your chatbot.

Don’t forget to design all bot content to match your brand voice, so you can provide a consistent experience across channels.

Here are a few examples of successful avatars:

#4. [ Facebook Messenger Only ] Plan for compliance with Facebook’s Messenger policy.

Facebook has rules for how businesses can communicate with users via Messenger bot. These rules exist to protect users from spam, and to make sure they get maximum value from Messenger.

Before building, get to know these key principles from the Facebook Messenger policy. When you design your bot, make sure it complies.

(And if you plan to use reengage messages and/or sequences in your bot, you should also look over Facebook’s most recent changes related to these two functions, in effect as of March 4, 2020, as well as the December 16, 2020 changes in response to new EU privacy laws).

Conversational AI glossary

Before jumping right in, let's cover some frequently used terms in conversational AI. We've put a list together of terms with their definition so you can navigate the app with ease.

Conversational AI aka Chatbot (or bot for short)

A conversational AI is software that imitates human-like conversation with users via websites and mobile messaging apps. They answer users’ questions and help them handle tasks without the help of a human.

Bot Studio (formerly called authoring tool)

Bot Studio is our feature that allows you to build and launch chatbots with little or no coding.

Conversational channel

A channel is a place where a chatbot is launched and where users can interact with it. It could be a website, LiveChat, Facebook Messenger, WhatsApp, Slack, Kik, etc.

Artificial intelligence

Artificial intelligence (AI) is the area of computer science that is sometimes described as machine intelligence. It allows conversational AIs to perform tasks that would normally require human intelligence, like decision making, language translation, or speech recognition.

Generative AI

Generative AI refers to artificial intelligence systems that can generate new content, including text, images, music, and other media, by learning from vast datasets. These systems utilize algorithms to understand patterns, styles, or structures in the data, enabling them to produce novel outputs that are similar to, but not exactly the same as, the examples they were trained on, often achieving results that can mimic human creativity.

Natural language processing

Natural language processing (NLP) is a branch of AI that helps computers read, understand, and make sense of the human language. It lets chatbots understand the context and meaning of a user’s message and react to it properly, even if the user makes a spelling mistake.

Apart from chatbots, NLP powers plenty of apps and services including Google Translate or Grammarly.

Machine learning

Machine learning (ML) is a branch of AI that allows chatbots to identify patterns in human language and learn from past conversations. Thanks to machine learning, properly programmed AI chatbots can improve over time without the help of a human.

Sentiment analysis

Sentiment analysis is a subfield of computer science that uses NLP and machine learning to measure the sentiment and tone of a text or spoken language.

Sentiment analysis can help a conversational AI analyze user messages and identify whether the person’s attitude towards certain products or services is negative, positive, or neutral.

API

API is the acronym for application programming interface. It’s a toolset that lets apps communicate with other services and enables developers to integrate new apps into existing software.

For example, when Facebook made its API public in 2016, chatbot developers were able to finally integrate their third-party solutions with Messenger.

Webhooks

Webhooks, also called a web callback, are automated API responses. They retrieve information from chatbot conversations, like email addresses or telephone numbers, and pass them automatically to web services. For example, to your CRM or email marketing software and the other way around.

Software integration

Software integration (or just integrations) is the process of two or more applications being connected. For example, with Messenger integration, you can connect ChatBot with Messenger or Slack with no coding.

Entities

Entities are keywords and phrases that are similar in meaning and that are crucial for your chatbot conversations. For example, the names of products and services you offer.

You can create your list of entities so that your chatbot can identify and retrieve them from user responses. Because of that, your chatbot can provide answers relevant to user input. For example, it can recommend an exact product that interests a customer.

User: I would like to order coffee. Bot: What type of coffee would you like to have?

Latte
Flat white
Cappuccino

Chat widget

A chat widget is a ready-to-use, customizable chat window that is added to a website. A chat widget hosts your chatbot and enables users to converse with it.

Groups (or Flow)

Groups are collections of related topics, like "Check My Insurance" or "Find a Location," which help organize your chatbot’s conversational topics into manageable sections.

Greeting

A greeting is a short message that welcomes website users and encourages them to start a chat. It’s shown right above the minimized chat widget.

Fallback

It’s a default interaction that is triggered when the bot doesn’t recognize the user’s input.

I missed what you said. Could you repeat it?

Template

It’s a feature that allows you to make a copy of an existing chatbot scenario. Templates are helpful when you need to build a complex chatbot flow but you don’t want to start it from scratch.

Context

Context helps your chatbot decide which response should be sent to users based on the context and the information the user provides.

For instance, your chatbot can ask a user if they are looking for a covid test or schedule an appointment, and then display an offer relevant to their response.

Context lets you also show specific info to people who come to your website using a specific URL address. This comes in handy if you want to personalize your chatbot communication.

Contextual codes (formerly known as referral codes)

Contextual codes let you show specific info to people who come to your website using a specific URL address. This comes in handy if you want to personalize your chatbot communication.

Cards

Cards are a one of the most important functionalities of the app. They're located in the bot builder and they allow you to build your flows and showcase your content. Located at the bottom or you can click on the + sign on the right inside the block.

Now let's dive into the different types of cards:

Text card

The text card is your best friend! You will be using this card the most and that's because a conversational AI is mostly well conversational 😉 This card will be used to share text content with your visitors. Some common use cases include: answering FAQs, asking a question, describing next steps, and much more.

Image card

The image card is practical when you want to share an image with the visitor. It could be a product image or even a map.

Quick reply card

Quick replies are short chatbot messages that suggest the visitor a few pre-set possible options. They help users make their decision faster and improve the flow of the conversation.

Carousel card

The carousel card is useful when you need to display to the visitor a few options but in a richer format with more information (image, text, button, and link).

Handover card

Handover is an action that redirects a chatbot user to a live chat agent. It can be triggered when the user’s problem goes beyond the chatbot’s scope.

Delay card

The delay card is a typing delay that enables you to set up how fast your chatbot should respond to user input. It lets you create a more natural flow of conversation. It's particularly useful when you have a few text heavy messages in a row and you want to give the visitor a chance to read every message without being bombarded.

Video card

The video card operates similarly to the image card. It's useful when you want to share rich content with the visitor. It could be a video tour of your facility, a tutorial or anything else that can provide value.

API card

We quickly went over the API definition and this is how you can leverage that technology. The API card allows you to connect to your systems whether it's a CRM, EHR, Yext, insurance coverage, and much much more. This card is the most powerful tool in your arsenal. It allows you to show content dynamically and at scale to personalize the experience.

Attribute card

Attributes can be described as packages including information (e.g., name, email, phone number) a chatbot collects while chatting with a user. Attributes help a chatbot personalize its responses.

Bot: What’s your name? User: Jacob. Bot: Thanks! How can I help you, Jacob?

Go-to card

The Go-to card works in sync with the attributes because it allows you to set custom rules based on attributes.

Let's say a visitor is scheduling an appointment with your conversational AI and in that flow there is a fork (a split into two flows). If the visitor is older than 70 years old or younger than 70 years old, the flow is different because you need to ask an additional question for visitors above 70. If the visitor selects 70+ then you can tag him with an attribute and have the AI ask him the extra question. If the attribute is not populated, the visitor will skip that question.

SetAI card

The SetAI Card allows you to dynamically set attributes in your chatbot based on a prompt that you provide. These attributes get set in real-time on the runtime of each session with your chatbot.

Got all that? Now you’re ready to build your bot! 🛠️

← Previous

Quick Start

The Web App