Chat bots have come a long way from only being the customer service representatives. Imagine this; they can be even your kid’s new online friend, his or her BFF (Best Friends Forever). Yes, you read that right.
Mitsuku, a chat bot created by Steve Worswick took home the Loebner Prize in the year 2013. This was in an annual competition that identified world’s most human-resembling computers as well as artificial intelligence programs – which could meet or even surpass the Turing test of human behavior. She can play games and do magic tricks at the user’s request.
Coming into present scenario, Mitsuku is not the only AI-powered chat bot responsible for winning the hearts but in China, in the year 2016, the chat bot Xiaoice, who is like a hyper-chatty, seventeen year old girl, is proven to be empathetic and engaging to over 10 million Chinese teens who have proposed to her over the chat platform Weibo by saying, ‘I love you’. This is in fact, a huge appreciation to the corporate parent of the bot Xiaoice, Microsoft.
As the two chat bots Mitsuku and Xiaoice are gaining popularity as well traction among the kids, there is a huge probability as well as a huge opportunity can be seen for the chat bots In the enterprise – the 24/7 customer service. Still not satisfied? Check out below mentioned reasons as why China loves Xiaoice.
Reasons why China loves Xiaoice:
- Xiaoice derives all the knowledge from the Bing search engine.
- She can engage in multiple conversations during one session.
- She can even tell jokes and up to date with latest pop culture gossip.
- This is not all; she also argues and disagrees if you are factually wrong. So, it’s time to get your facts right before engaging with Xiaoice.
- Lat but not the least, she also remembers your previous conversations. Approximately, about 26% of her database comes from the prior conversations.
Lauren Kunze, principal at Oakland, Calif-based chat bot platform Pandorabots which was actually used by Worwick to build Mitsuku throws some insight on the future of chat bots as well. She says that consumers should soon expect deeper conversations or connections with these human-like bots or humanoids of all kinds – from buying a service to solving a problem.
Lauren Kunze, who engages with over 235,000 developers to develop nearly 300,000 bots, says that current engagement is more with the chat bots is exponentially higher than with the service bots. Her findings also conclude that if the consumer engages with a service bot, even if it offers a small talk, they return to it more often than not.
The studies also claim that the service bots should be excellent conversationalists rather than being only the task oriented robots designed to solve specific problems. It will help to create an engaging and helpful customer experience for some of the biggest companies of the world.
A friend in need, indeed!
Unlike the service bots created by Apple like Siri and one by Amazon, Alexa often pride themselves on completing tasks, Mitsuku and Xiaoice just want to know how you’re doing. They even remember who you are and what you like. A great thing about them is that they respond to you more appropriately, empathetically to your expressions whether it’s remorse or elation. This is the reason why millions of people love engaging with these service bots to get things done.
It is believed by Steve Worswick that the same logic will be applied to future or next generation of customer service technologies. For example, Software giants like TeleTech along with other software technology companies are already heading in this direction which is also agreed by Lauren Kunze. All the difference, in fact is made by just a small talk.
In her own words, “People are now expecting human-like interaction with bots,” she says. “So the content these bots create has to evolve. An example could be like the problem faced by the bots when the customer go off the topic. For this, developers have started implementing conversational tidbits into service bots leading to a result which is very positive.
With the fast advent of chat bots or the service bots to improve the customer experience, some of the major problems are faced by many companies in the customer service space. Firstly, they believe that the AI-powered bots are still not intelligent enough.
Secondly, bots are still to learn and understand natural language. There is a huge scope of work to be done in this area.
This can lead to switching the unresolved query to new bot – think bot swarm or a human agent, which is quite poor or even non-existent at the time. This leaves the customer’s query unresolved. As a result, it’s poor customer service.
A lack of systems integration can be largely covered by a human agent but it cannot be covered by a bot. A bot can only depend on constant indexing of the corporate knowledge.
Dealing with these problems is the challenge every company which is implementing service bots or chat bots has to face.
How to get started then?
The question is still daunting. But there are certain things to be considered before jumping into providing service bot or chat bot based service to your loyal customers. Benefits can go beyond the cost reduction like business benefits by improved customer satisfaction through reduced response times, increasing customer convenience and increasing first time resolution to their problems.
The first step is to understand the bots in every aspect, not alone in your industry but others as well. There are several examples of bots being used in food and travel industries and how they use different interfaces to interact with customers.
The next step you could take is to understand and analyze the potential of using a service bot in your own organization. Like if you are already using means like messaging or chatting to solve your customer’s problems, you can incorporate a chat bot to do the job for you. If you are not using any such means, you can analyze social forums, call logs or email histories to see what all problems could be solved by using chat bots.
Lastly, run a pilot with the human agents as well as chat bot side by side to test and train it to handle more complex scenarios. This will enable you to test what type of interactions work best with your customers.
Share this article and follow me on Twitter. I would love to connect with you.