What is a chatbot?
Have you ever navigated a commercial web-site without any luck on finding what you were looking for? Or maybe you were using your smartphone to this end but the page was not visualizing well? Or also, for sure you’ve been in the situation were you were asked to download a commercial app (e.g. the BurgerKing app or your regular shopping mall’s app) but you didn’t, because you were lazy or just had not enough free space in your phone.
Is there a direct solution to these problems? Is there a way to digitally access to these commerces in a more natural way? What would you think if you could address a virtual assistant to attend your needs? This is what chatbots is about.
A chatbot is a service which allows us to interact with a business through a chat, as it can be deduced from its own name. For a start, you only have to talk to the bot using speech or text messages (depending on the interface), the same way you would address to a person; that is using Facebook Messenger, Slack, Telegram, etc.
Chatbots can be useful in a number of situations, being able to solve functional issues (e.g. a bot that guides you to buy the best products in an e-commerce), but also can be of help regarding your leisure time (e.g. a bot that advices you on what films you should go to the cinema to watch).
How does it work?
Ok, we can communicate with a chatbot through questions and queries, talking to it as if it was an actual person. But, how does the bot to understand what you say and answer you properly?
The short answer is that the chatbots have been programmed to analyze your inputs. This is what we call ‘artificial intelligence’. But to be fair, we should tell appart two different kind of bots: some that are not so ‘intelligent’, based on a set of predefined rules, and others that use Machine Learning (ML) to achieve certain degree of intelligence.
The chatbots that are based on rules are quite simple. Their behavior is similar to that of the phone bots, the ones that guide you through spoken menus (“Dial 2 if you call regarding this or that”). With this kind of bots, the answers are automatized. If your question or query is beyond its rules, it won’t be able to answer you, as they lack of flexibility.
On the other hand, the chatbots that use ML can deal with a wider range of situations, far better than the rule-based bots. This bots have an ‘artificial brain’, being capable of understanding natural language, but also understand the context of the conversation, adapt to it, and finally can learn from the users to guide them more effectively. In fact, the great advances in ML during the last years have given place to what would be for sure then next great revolution.
How does a chatbot fit with my business?
Why should I consider to deploy a bot for my business? How do I know if the people is going to use it?
Believe it or not, having a commercial bot is today a synonym of opportunity. And it is not us telling this, its the big numbers. The global trends clearly show that the people is using messaging apps more and more, having surpassed even the numbers of the usage of social networks. This combines with the tendency to install less mobile apps.
And what is a business about? It is about finding the right people, wherever they might be, to offer them a solution to their needs and problems. And, being honest, currently the people is mostly in one place: the messaging apps. The best of all this is that they won’t have to install anything nor even access your website. The users will be in contact with your business by means of their regular messaging app.
Chatbots will improve your business in many ways, by offering a better service to your clients. They will also help you sell more and more efficiently (e.g. by showing the right products to the right clients) and given the natural way of interacting with the bots, this will increase the customers interest in your company.
So do not hesitate, deploy your bot and help your clients, by recommending them what to buy or just by giving them assistance. The range of opportunities is immense.
How is a chatbot developed?
The good news is that deploying a chatbot is currently feasible for any company. There are a bunch of tools and growing community of experts that will guide you through the process of imaging and setting up a bot they way you would like to. Actually, beyond the technical challenge, most of the difficulties you’ll find will be of concept. That is, you will have to decide how your bot should behave and interact with your clients. This will be the key to offer a unique and attractive experience that will engage your users.
At a technical level, the big players (Google, Facebook, etc.) have invested heavily on this market, offering open tools to provide your bot with the intelligence you’d like. We can remark two platforms in this sense: api.ai, from Google, and wit.ai, from Facebook. Both offer a friendly environment to create a bot, train it to interact with your users and finally embed it with the most common messenger platforms. However, there are other platforms also providing similar services, so you can take the one that adapts better to your needs: Octane AI, Howdy’s botkit, textit.in, Motion.ai, Chatfuel, IBM’s Watson, BeepBoopHQ, Betaworks’ Dexter, converse.ai or Gupshup.
Are you ready?
If you’re considering deploying a chatbot for your business it would be a good idea to contact with experts to guide you. In Tyris Software we can give you advice on how to focus your bot, and we also cover all the phases: development, training and optimization. We have a great team with years of knowledge in Machine Learning and ready to share our ideas with you to shape yours. Our goal is to make this technology a success for you.
So, do not hesitate and invest in your future! If after all you are still thinking about it, ask a chatbot to see what it thinks