Chatbots — who’s doing it well, and how are they doing it?
Search

Chatbots — who’s doing it well, and how are they doing it?

By Aditi Kant


While chatbots are relatively new, you might not know them as you do today without Alan Turing’s famous Turing Test, established way back in 1950. If you’re not familiar with the Turing Test, it’s a test of a machine’s ability to exhibit intelligent behaviour comparable to, or indistinguishable from, that of a human. With its creation, Alan Turing unknowingly established the foundation for chatbots. Cool right?


More recently, Facebook’s messenger service has become a widely known platform where bots interact directly with Facebook users. It may seem like messenger has been around forever, but it was only launched in 2016. Since then, there’s been no looking back. Chatbots seem to be everywhere and many companies have adopted chatbots to personalise their business operations without having to hire large customer-facing teams.


According to eMarketer, 63 per cent of surveyed customers said they were more likely to return to a website that has a live chat option. According to Gartner, “By 2019, 40% of enterprises will be actively using chatbots to facilitate business processes and natural language interactions. By 2020, customers will manage 85 per cent of their relationship with a company without interacting with a single human.”


While chatbots are taking the industry by storm, no company wants to be left behind. In this article, we will establish critical success factors for user’s experiences with chatbots and we will define some critical UX attributes that a chatbot must have to ensure the users get an optimum experience.


So what makes it a great chatbot? The first and most important requirement is to make a chatbot omnichannel. It should be available across all of a services digital platforms, including mobile, web and application, to name a few. It is critical that users experience continuity in their conversations as they move across the platforms during a conversation.


A flexible chatbot allows users to interact freely with the chatbot in a way that suits them best. The inputs should allow both text and voice interactions. Depending on the user’s preference, they choose the appropriate medium and the chatbot should be intelligent enough to understand the inputs and give proper responses.


It is not sufficient to give back relevant content unless it is also in an appropriate tone. The tone and sentiments of the users should be responded to. For example, an angry message needs an appeasing response and not an aggravated one. Another factor that plays a role in phrasing the responses is a brand’s values. These should always be consistent in all customer interactions.


Providing feedback to users keeps them engaged and aware of what is happening. It’s not just true about the Graphical User Interface (GUI) but also for the Conversational User Interface (CUI). Provide messages at the outset of a conversation to set users’ expectations and keep giving real-time feedback when the bot is preparing to answer (eg. an animated typing bubble). When an error is made, feedback on how to fix the mistake should be provided immediately.


According to Mobile Marketer research, around 40% of millennials say they chat with chatbots on a daily basis. This means we still have a huge population who needs to adopt the technology with comfort and trust. A chatbot’s last but most critical UX attribute is to generate trust among users. Trust is generated when a chatbot can honestly introduce itself as a bot. At times when AI fails to provide the right content at the right time and in the right tone, if the users can be told that their conversation will be handed over to a “real” person, it will only make them more trusting. People appreciate honesty.


Here are examples of chatbots from the region that we evaluated against the above-mentioned attributes.



Findings

Through this review of the chatbot apps offered by some big businesses in the region, it is clear that the full potential of chatbots has not been tapped into. The above-reviewed apps do not efficiently support key user tasks.


The apps are primarily used for customer service purposes. We could not find any of the retail giants offering e-commerce possibilities to users through bots on Facebook Messenger, Slack or WhatsApp. Simple tasks like booking a hospital appointment can be done by chatbots but there were no examples or guidance to be seen. For most purposes, companies are pushing users to use a call centre or other means of assistance.


Successful chatbots should follow the above-established best practices when it comes to defining a positive user experience. A chatbot should mimic the highest level of support offered by the company, and ensure that the user experience is seamless. Essentially, chatbots currently have a long way to go but a ton of progress is being made in this area with regards to humanizing chatbot interactions.

0 views

Follow us here

  • LinkedIn
  • Grey Instagram Icon
  • Grey Twitter Icon
  • Grey Facebook Icon
  • Meetup
  • Grey YouTube Icon
  • Medium Logo_edited

Subscribe for Digital of Things communications. 

By subscribing, you are agreeing to our privacy policy.

Call us

Email us

Visit us

1504, Indigo Icon, Cluster F,  

JLT, Dubai, UAE.

Copyright © 2018 - Digital of Things DMCC