Messaging has become one of the leading ways we communicate. Sending voice or text messages is fast, it’s convenient, and lets us connect with others whenever we want.
Interestingly, these instant methods of communication have made us less patient. Our attention span has shortened, and we get frustrated when we can’t seamlessly connect with friends and family — and the same happens when we try to connect with brands.
New communication trends have made businesses realize that to keep up with rising customer expectations, they need to adapt and transform their customer service. Living up to the changing expectations of “always-connected” customers is neither easy nor cheap, but there’s a solution that can help brands meet them, and that solution is chatbots.
Introduction to Chatbots
Welcome to Chatbot Academy, a complete course that provides an overview of chatbots and will teach you how to build them from scratch, without coding.
I’m Kaia with LiveChat, and together with our chatbot experts, we'll walk you through your journey to master chatbot building.
Since you’re already here, I’m guessing you’re a business owner who’s considering adding chatbots to your customer service armory. Maybe you’re thinking about starting a new career as a chatbot designer, and you need to level up your skills. Or, you just heard about chatbots and you want to find out if they’re worth exploring.
If you belong to any of these groups — you’re in the right place. We created this course to give you a beginning-to-end understanding of chatbots and how they work. During the lessons, you’ll gain practical skills with which you can plan and design successful chatbots using a no-code platform. You’ll also learn how to implement and optimize chatbots with ease.
Let’s dive in.
What’s a chatbot?
I assume you already have some knowledge of chatbots, but let’s organize what you know. We’ll start by answering the pillar question — what exactly is a chatbot?
The most general definition of a chatbot says: it’s computer software that can have a conversation with a human in a natural language via a website or a messaging app.
Let’s explain this with an example.
Imagine you visit an online food store that provides chatbot support. You notice a little chat bubble at the bottom of the page and when you select it, you can start asking the chatbot questions, like:
“How can I return my purchase?”
“What products do you sell?"
“What is my order status?”
and the chatbot provides you with answers immediately.
Or alternatively, you visit the same store, not noticing the chatbot. You start browsing some offers, and after two minutes you get a proactive greeting informing you, for instance, about the latest promo offers.
Now you may think — that’s basically what a human can do. That's right, but a human can’t effectively chat with multiple users at the same time — a chatbot can, and it can do it around the clock without getting out of breath.
And, by the way, don’t confuse chatbots with bots. A bot is a type of software that can do repetitive tasks, but it can’t chat with users.
We can distinguish between, for instance, search bots that are used by Google to index the content of the web, and feed bots that can look for news on the web and add it to news sites.
Bots can also be used for malicious activities. They can generate fake clicks, break into user accounts or steal data. Using malware software is cybercrime and should be avoided at all costs.
Now that you know what chatbots are, we’ll dig deeper into how they function.
Two Types of chatbots
There are two ways a chatbot can understand what humans say, and they can be divided into two groups — AI chatbots and rule-based chatbots.
AI chatbots can understand human language, its context and its meaning, and they can mirror that behavior. They apply
natural language processing, which lets them understand how humans communicate
and machine-learning, which allows them to learn from past conversations with users.
So, the more an AI chatbot communicates with users, the more patterns, words, and contexts it learns. This knowledge lets them freely create answers to more questions.
However, this characteristic turns out to have a drawback. AI chatbots reflect the users’ behavior, but as we can’t control what users say, we can’t fully control what an AI chatbot learns or what answers it provides.
This aspect is a challenge for designers and developers because it’s still difficult to create an AI chatbot that can deliver a consistent experience and predictable answers.
What are rule-based chatbots?
Because of that drawback, more businesses use rule-based chatbots, which are simpler — they can be built by using a platform — cheaper, and more controllable.
As the name suggests, rule-based chatbots function on applied rules. This means they only answer questions they were prepared to answer by using predefined answers.
How does this work in practice?
Let’s say you run an online store and you want to inform users about your delivery costs by using a rule-based chatbot. To do so, first you need to train your chatbot on how customers may ask for a delivery price — meaning what words and phrases they might use. Then, you have to teach your chatbot the exact answer you want it to provide for these questions.
However, if your customer asks the chatbot a question like “Which color is better, yellow or black?” and you don’t train it to answer this question, it won’t help the customer. Rule-based chatbots can't freely create answers to questions they don’t know. In that case, the chatbot can, for example, transfer a chat to a human agent so they can resolve that query.
With this in mind, we can say that rule-based chatbots aren’t as advanced as AI chatbots. But surprisingly this is their advantage. Although they can’t freely answer all the users’ questions, their behavior can be designed from A to Z. You can use a rule-based chatbot to help customers solve certain types of problems and be sure that every customer receives the same answer and a consistent experience.
There are many areas a chatbot can help a business, starting from customer support.
Chatbots in Customer Support
Chatbots can work alongside your support agents or as a standalone, after-hours service. They can help the customer solve common problems, book appointments, or schedule calls.
For instance, at LiveChat, we handle lots of chats daily, and answering them has always been an effort. At some point, we decided we were gonna build a chatbot to help our support team deal with all those inquiries, and so we did.
Our chatbot soon bolstered our frontline customer support. We prepared it to filter spam messages and resolve common issues. We noticed that within the first three months after the launch, the chatbot closed 36,886 cases without the need to transfer visitors to human agents. We also discovered that it was able to take care of 28.6% of our live chat communication.
Chatbots in Marketing
Brands use chatbots to enhance their marketing activities, too. The chatbot can invite a customer to chat and collect their contact details. It can share coupon codes, provide personalized product recommendations, and promote your job offers.
Take, for example, the Valley Driving School, a Canadian driving school. The brand uses their chatbot, Mr. Ro-BOT-o, to proactively engage users. The chatbot promotes their courses, answers regulatory questions, and engages leads who’re considering purchasing a driving course — it schedules their training — or passes their chats on to a live agent.
By adding a chatbot into their services, the company was able to reach a 94% customer satisfaction rate.
Chatbots in Sales
Chatbots can be of great use for sales teams as well. They can help acquire new leads at lower costs. For example, Toyota Poland uses a chatbot to initiate the sales process with a prospective customer. The chatbot asks the user a set of questions, gets the user’s details, and passes them on to a selected car dealer, who based on collected information, provides the potential customer with a sales offer and a financial calculation.
This way, the chatbot can generate leads even after the sales team has packed up and gone home. By adding a chatbot to the sales process, you can spend less time on acquiring leads and focus more on building relationships with prospects.
That’s all for this lesson. Let’s sum up what we’ve just learned!
A chatbot is computer software that can chat with the user via a computer or messaging apps.
We can distinguish between rule-based chatbots that can answer questions by using predefined answers and AI chatbots that can interpret human language and communicate with users more freely, but the latter are more difficult to control.
Chatbots help businesses improve many processes — unload the customer service team, enhance marketing activities, and support sales teams.
In the next lesson, we’ll talk about the ways you can build a chatbot for your business. See you soon!