What makes a great travel chatbot?

Part 3: It’s a new era!

2017 was advertised as the year of the chatbot and 2018 is already going into the same direction. In fact, AI-powered bots have by now proved their merits and value when it comes to bringing in business and providing customer support! Both of these are factors that the travel industry heavily relies on. There’s no doubt that travel agencies, hotels or OTAs are players in a very competitive field, therefore they must make sure that they have the best offers and services in order to retain the most customers. With the automated travel assistants, they can do that and much more!

Chatbots are a cleaner alternative. By integrating on a website or a messaging app, bots do not create bloatware on your clients’ device. That is, there is no need to download several apps, instead, travellers are able to receive notifications, updates and promotional offers in just one platform! The travel industry has offered us some good examples of what chatbot technology can do. Some great solutions are already on the market. On today’s post I’m going to outline some features that you ought to make sure your AI assistant has:

Problem-solving is a must

A chatbot ought to solve its users’ problems otherwise is going to fail. Your bot must make it easier to get something valuable that existing methods can’t match. If not, travellers have got no incentive to use it over platforms they already trust and are familiar with. A good example is Mezi which is designed to be an all-in-one travel agent. The user tells him where they want to go and when, plus any additional requirements they might have (for instance, food allergies) and the bot does the rest. So, rather than using different platforms to find flights, hotels and other relevant information, travellers can do everything in one place.

Personalisation is key

Aside from being convenient and solving problems, chatbots should have the ability to customise search options and personalise the travel experience. Indeed, personalisation is all over the place — we experience it every single day, from personalised playlists on Spotify to recommended items on Amazon. Our online experiences have become custom-made to our personal interests and buying profiles. Being so used to it, makes us expect personalised treatment in every single aspect of our life. This is one of the ways that bots will become truly unique and useful for travellers! By mixing travellers’ preferences with data about a destination allows the bot to provide pertinent and contextualized information to its user. Its known that travellers are a different breed of consumers, they do not just want personalisation, they also want unique travel experiences. With a chatbot, you can make sure that this happens!

Be a fortune-teller

The fact that AI-travel assistants cannot anticipate users’ worries in the same way as a human travel agent can is a possible problem. In this sense, your bot should be as predictive as possible in order to be useful for users. It should make suggestions that continually drive the conversation forward and even help travellers discover other features of the bot. For example, if a user has booked a flight, the bot could recommend the user to book a hotel room in the area. Being predictive comes down to having people in your team brainstorming a variety of scenarios and users’ possible responses or desires.

Manage complexity

You should try to create a travel bot that can handle complex inputs! Ideally, a user should be able to ask for a one-way ticket to Budapest, Hungary from Schiphol airport on May 5, 2018, and the chatbot should handle all the request at once. However, technology is not at that point yet and a bot can’t handle every situation as well as a human. Consequently, you need to make sure you have a well-designed exit strategy! The travel bot has to understand when users are getting frustrated or when it does not have an answer to the query. In addition, the bot should also know when is the right time to handoff to a human agent.

Give multiple choice responses

Since natural language processing (NLP) and AI are still technological advancements in progress, most chatbots work with a decision tree format. Giving multiple choice responses often leads to higher engagement rates! As a matter of fact, it is known that travellers do not want to type every single answer, consequently, you ought to make sure that your bot has anticipated what they need and presented it when they actually need it. Providing multiple choice responses is almost a requirement these days!

It should be fit!

Just like us, Chatbots need to be trained regularly so they are shaped to tackle the daily queries from the users. So, it’s necessary to assist the chatbot to practice and learn, so he can make sense of the queries he did not understand in the past and, in the future, be ready to reply to similar messages. Let’s look at HiJiffy’s training feature to better understand what it means to train your bot. Well, it can be easily described as a classroom where every variable (which comes from different sources) is readily available, labelled and organized, so you can see the message in the right context, what is the intention of the message, and based on the intent, what is the right answer to give to the user — and you have access to this information all in one place! The more you train the chatbot, the more independent and smart he will become.

And you? What do you think makes a great travel bot? Comment below, I am curious to know your opinion!

This is the last article of a three-part series based on our upcoming webinar: How chatbots are changing the travel industry. You can find part 1 and part 2 here.

Our webinar has a limited number of attendees, if you have not registered yet, now is the time! We really hope you can make it, but if you can’t, register anyway so we can send you the recording. Here is the registration link: https://my.demio.com/ref/B6Hm8pf1x3VMa1qC

NB: This is a viewpoint by Mariana Marques from HiJiffy. HiJiffy brings chat to the hospitality and travel industry, enabling customers to effectively connect with you through instant messaging apps.