Over the last couple of years, the global interest and attention in chatbots has flourished. More and more people are talking about it every day. Huge companies like Google and Amazon started to introduce their own smart speakers: Google Home and Amazon Alexa. Other tech companies, like Facebook, Slack and Twitter, are introducing chatbots on their platforms. Chatbots are a new trend and companies are investing lots of time and money in them.
Chatbots are really interesting, and in this blog post I will talk about what they are, how they work, why you need them and how to build them.
First things first: Conversational interfaces and other used terms
Various terms are used to refer to chatbots and many terms are used interchangeably. One of the terms is a conversational interface. It is true that a chatbot is a conversational interface. But not every conversational interface is a chatbot.
A conversational interface is much broader than a chatbot. The official definition is “the front-end to a chatbot or virtual personal assistant, allowing the user to interact with the app using speech, text, touch and various other input modes”. Just like real humans can communicate in various ways, there are also various forms of conversational interfaces. In total, there are four:
- Spoken Dialogue Systems
- Voice User Interfaces
- Embodied Conversational Agents
Spoken Dialogue Systems (SDSs) and Voice User Interfaces (VUIs)
Spoken Dialogue Systems (SDSs) allow humans to interact with a computer on a turn-by-turn basis using spoken natural language for input and output. These systems were mainly developed in academic and industrial research laboratories.
At the same time, similar systems were being developed by companies, called Voice User Interfaces (VUIs). So, in essence, Spoken Dialogue Systems and Voice User Interfaces are really the same, but differ by who they are developed. The one by academics, the other by businesses.
Embodied Conversational Agents
Embodied conversational agents (ECAs) are computer-generated animated characters that combine facial expression, body stance, hand gestures, and speech to provide a more humanlike and more engaging interaction. An example of an ECA is Hello Barbie, a commercially available conversational toy that responds to children’s inputs and retrieves answers from data sources on the web. Here is a video of a CNET reporter talking to Hello Barbie:
What is a chatbot? A definition
Various definitions are used for chatbots. What all these definitions have in common is that there is a simulated conversation between a human and a computer. The definition I use is:
A chatbot produces simulated conversations in which the human user inputs something and the chatbot makes a response
In other words, chatbots facilitate a conversation between a human and a computer. Usually, you have a conversation with a friend on your messenger app, for example Whatsapp. But with chatbots, it is slightly different. Because you do not talk with another human, but with a computer.
The difference between chatbots and the other terms mentioned above, is that chatbots are more limited. Most of the time, chatbots are displayed on a screen and you can only use text and sometimes voice to communicate with them.
An example of a chatbot…
EpicReads is a good example of a chatbot. It advices book lovers which book they should read. The customer can select a book/author he loves, or the mood he is in right now. Based on the customer preferences, the chatbot automatically suggests some books that the user could read. The chatbot is live right now, so if you have Facebook Messenger, you can try it out here.
How does a chatbot work?
A chatbot works relatively simple. I will explain it with an example:
- A user starts a conversation with a chatbot of a car company. He asks “Which red cars do you have?”
- The chatbot interprets the message and recognizes the words “red” and “cars”.
- Based on the recognized words, the chatbot will look into a database for “cars” that are “red”.
- Finally, the chatbot will send a message back to the user with all the red cars that he found in the database.
In general, most chatbots work this way. It is also possible that the chatbot does not understand the question and then he will respond with “I do not understand you, could you rephrase that question?”
If you want to know more about how chatbots work, and the more technical side of it, check out this post!
A bot is not the same as a chatbot!
Nowadays the terms bots and chatbots are used interchangeably. However, there is an important difference between those two terms. Bots are software applications that perform automated tasks. For example, a bot could be a software program that automates social media messages, or software that automates your accountancy.
A chatbot is also a software application that performs automated tasks, but differs in the fact that it also engages in a conversation with the user.
Where are chatbots used?
Nowadays, chatbot are mainly used on three places:
Often a chatbot is added to a website for asking standard questions, qualifying leads or referring the customer to the right department within the company.
- Social media platforms
Since social media platforms like Facebook and Twitter integrated chatbots in their platforms, a lot of chatbots have been created there. Chatbots are used for various purposes, from shopping products to emotional support to entertainment.
This is the least used option nowadays. Chatbots can be integrated in apps for asking standard questions, for asking how the app works or for getting help from customer service. Integrating a chatbot in an app is more difficult than on a website or a social media platform.
What can a chatbot do?
Chatbots are used for different reasons, but there are mainly three roles:
- Role of a digital assistant
With the role of digital assistant, the chatbot helps the user with doing certain tasks. They are mainly used to increase productivity and doing things faster.
- Role of an information provider
With the role of information provider, a user asks questions to a chatbot and it providers answers to these questions. They are mainly used for obtaining help and information.
- Role of a general chatbot
With the role of a general chatbot, a user uses the chatbot to conduct a conversation for entertainment or emotional support.
Why do I need a chatbot?
As said before, more and more businesses are implementing chatbots everyday. This is a good thing, because it adds value for your customers and for your business. But why should you use it for your business? Here are some reasons:
- 24/7 available, everywhere
With a chatbot, your company is 24/7 available for your customers. Wherever they are, whenever they want. Your customers can ask questions to your company, even in the middle of the night. According to Chatbots Magazine, this is one of the main reasons why customers use chatbots.
- Save on customer service costs
Chatbots can save a lot of money in terms of customer service. 80% of customer service questions are standard questions, which could be easily answered by a chatbot. The result? Save up to 80% on customer service representatives. And customers are willing to talk to chatbots, because research of Hubspot shows that people are really open to using chatbots, as long as they get help quickly and easily.
- Boost your sales
A chatbot can boost your sales tremendously. With a chatbot, you can sell directly to your customers via a chatbot. Another way, is to promote your products in a chatbot and lead them towards your website.
How to build a chatbot?
You can build a chatbot in two ways: build them from scratch or use chatbot software. The choice for using one of the two depends on the goal of the chatbot and the time/budget you have for the project. I will discuss the pros and cons of both alternatives.
Building from scratch
When you are building a chatbot from scratch, you need to code it. You could do this yourself (if you are a developer) or you could hire a software development company to do it for you.
- You can make it exactly how you want it
- Complete control over the design of the chatbot
- Make changes easily, according to changing needs
- It is more expensive than using chatbot software
- It takes more time than using chatbot software
You could also use pre-built chatbot software to build your chatbot. With pre-built software, you don’t need to have coding knowledge to build your own chatbot. Most of these software companies also provide (video) tutorials and templates on how to make your chatbot, which makes it even easier. Examples are Chatfuel, Manychat, Octane AI and Mobile Monkey.
- It takes less time than starting from scratch
- It is cheaper than starting from scratch
- Can make it without any coding knowledge
- Can implement it easily on many platforms
- Less flexibility
- You have little or no control about the design of the chatbot
- Sometimes they only provide one platform
People start to understand the value of chatbots. They can automate conversations, so you can:
- Drastically reduce your costs on customer service
- Increase your sales
- Automatically qualify your leads
This technology is still at an early stage and a lot of companies don’t use it (yet). By using a chatbot for your company, you are really standing out from the crowd.
Are you convinced to invest in a chatbot? Leave a comment!
A chatbot produces simulated conversations in which the human user inputs something and the chatbot makes a response. In other words, chatbots facilitate a conversation between a human and a computer.
A chatbot works relatively simple. I will explain it with an example:
1. A user starts a conversation with a chatbot of a car company. He asks “Which red cars do you have?”
2. The chatbot interprets the message and recognizes the words “red” and “cars”.
3. Based on the recognized words, the chatbot will look into a database for “cars” that are “red”.
4. Finally, the chatbot will send a message back to the user with all the red cars that he found in the database.
They are 24/7 available, can drastically reduce your customer service costs and can boost your sales.
You can build a chatbot in two ways: build them from scratch or use chatbot software. The choice for using one of the two depends on the goal of the chatbot and the time/budget you have for the project.